From vincent-magnotta at uiowa.edu Fri Jan 2 12:12:23 2009 From: vincent-magnotta at uiowa.edu (Vincent A. Magnotta) Date: Fri, 02 Jan 2009 11:12:23 -0600 Subject: [slicer-devel] Windows Slicer 3.3 Alpha Version Message-ID: <1230916343.6516.21.camel@ganglia> I have posted a version of Slicer3 version 3.3 that was built under Windows to be used for the Meshing tutorial on Wednesday. If anyone has time, please try to download and test if the version will run on your Windows system. The build is missing some standard command line modules because of compiled size limitations with Visual studio. http://mri.radiology.uiowa.edu/downloads/Slicer3-3.3-alpha-2008-12-28-win32.zip Thanks, Vince -- Associate Professor Department of Radiology 0453-D JCP 200 Hawkins Drive Iowa City, IA 52242 E-mail: vincent-magnotta at uiowa.edu Phone: 319-356-8255 Fax: 319-353-6275 Website: http://mri.radiology.uiowa.edu From pieper at bwh.harvard.edu Fri Jan 2 19:45:30 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 02 Jan 2009 19:45:30 -0500 Subject: [slicer-devel] Windows Slicer 3.3 Alpha Version In-Reply-To: <1230916343.6516.21.camel@ganglia> References: <1230916343.6516.21.camel@ganglia> Message-ID: <495EB52A.3010007@bwh.harvard.edu> Hi Vince - That zip file doesn't work for me - it looks like the vtk*dll files (like vtkCommon.dll etc) didn't get copied to the bin directory. Maybe if you build the PACKAGE target in visual studio? In any case will the latest nightly be good for the tutorial now that Katie has a build with some of the problematic modules stubbed out? http://www.na-mic.org/Slicer/Download/Nightly/Slicer3-3.3-alpha-2009-01-01-win32.exe -Steve Vincent A. Magnotta wrote: > I have posted a version of Slicer3 version 3.3 that was built under > Windows to be used for the Meshing tutorial on Wednesday. If anyone has > time, please try to download and test if the version will run on your > Windows system. The build is missing some standard command line modules > because of compiled size limitations with Visual studio. > > http://mri.radiology.uiowa.edu/downloads/Slicer3-3.3-alpha-2008-12-28-win32.zip > > Thanks, > Vince > From vincent-magnotta at uiowa.edu Sat Jan 3 10:52:15 2009 From: vincent-magnotta at uiowa.edu (Magnotta, Vincent A) Date: Sat, 3 Jan 2009 09:52:15 -0600 Subject: [slicer-devel] Windows Slicer 3.3 Alpha Version References: <1230916343.6516.21.camel@ganglia> <495EB52A.3010007@bwh.harvard.edu> Message-ID: <7121AA62B4C3B94285742839606915A60337DB48@HC-MAIL11.healthcare.uiowa.edu> Steve, The version Jan 1 will work for us. We will do some more testing here this weekend, but we have gotten it to work on a couple of machines. Thanks for your help. Vince ---------------------- Associate Professor Department of Radiology 0453-D JCP 200 Hawkins Drive Iowa City, IA 52242 E-mail: vincent-magnotta at uiowa.edu Phone: 319-356-8255 Fax: 319-353-6275 Website: http://www.radiology.uiowa.edu -----Original Message----- From: Steve Pieper [mailto:pieper at bwh.harvard.edu] Sent: Fri 1/2/2009 6:45 PM To: Magnotta, Vincent A Cc: slicer-devel at bwh.harvard.edu Subject: Re: [slicer-devel] Windows Slicer 3.3 Alpha Version Hi Vince - That zip file doesn't work for me - it looks like the vtk*dll files (like vtkCommon.dll etc) didn't get copied to the bin directory. Maybe if you build the PACKAGE target in visual studio? In any case will the latest nightly be good for the tutorial now that Katie has a build with some of the problematic modules stubbed out? http://www.na-mic.org/Slicer/Download/Nightly/Slicer3-3.3-alpha-2009-01-01-win32.exe -Steve Vincent A. Magnotta wrote: > I have posted a version of Slicer3 version 3.3 that was built under > Windows to be used for the Meshing tutorial on Wednesday. If anyone has > time, please try to download and test if the version will run on your > Windows system. The build is missing some standard command line modules > because of compiled size limitations with Visual studio. > > http://mri.radiology.uiowa.edu/downloads/Slicer3-3.3-alpha-2008-12-28-win32.zip > > Thanks, > Vince > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090103/146a929f/attachment-0001.html From curtislisle at knowledgevis.com Sat Jan 3 11:14:46 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Sat, 3 Jan 2009 11:14:46 -0500 Subject: [slicer-devel] Windows Slicer 3.3 Alpha Version In-Reply-To: <1230916343.6516.21.camel@ganglia> References: <1230916343.6516.21.camel@ganglia> Message-ID: <92998EB9-CE41-4159-9395-20E19EF00518@knowledgevis.com> I have also tested the Win32 version of Slicer3 with IA_FEMesh from the nightly build from 01/01 on the following systems: 1. XP Service Pack 2 32bit on Pentium D with 1Gig RAM - works fine 2. Vista Home Basic 32 on Intel Core2 6700 with 4Gig RAM - work fine 3. Vista Home Basic running on Parallels over Mac OS X 10.5 on a powerbook - works except assigning materials on Load/BC menu. I wasn't able to get the rubber band widget to select nodes for materials. I am not too worried because of the funny Windows-over- Parallels-over-Mac configuration. The Mac binary version of Slicer3-alpha-01-01 works fine, including the node selection widget. So any Mac users can use the Mac binary during the turorial. Curtis Lisle curtislisle at knowledgevis.com On Jan 2, 2009, at 12:12 PM, Vincent A. Magnotta wrote: > I have posted a version of Slicer3 version 3.3 that was built under > Windows to be used for the Meshing tutorial on Wednesday. If anyone > has > time, please try to download and test if the version will run on your > Windows system. The build is missing some standard command line > modules > because of compiled size limitations with Visual studio. > > http://mri.radiology.uiowa.edu/downloads/Slicer3-3.3-alpha-2008-12-28-win32.zip > > Thanks, > Vince > > -- > Associate Professor > Department of Radiology > 0453-D JCP > 200 Hawkins Drive > Iowa City, IA 52242 > E-mail: vincent-magnotta at uiowa.edu > Phone: 319-356-8255 Fax: 319-353-6275 > Website: http://mri.radiology.uiowa.edu > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From fedorov at bwh.harvard.edu Sat Jan 3 19:40:51 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sat, 3 Jan 2009 19:40:51 -0500 Subject: [slicer-devel] Windows Slicer 3.3 Alpha Version In-Reply-To: <92998EB9-CE41-4159-9395-20E19EF00518@knowledgevis.com> References: <1230916343.6516.21.camel@ganglia> <92998EB9-CE41-4159-9395-20E19EF00518@knowledgevis.com> Message-ID: <82f536810901031640l42e900a0j7ae25db71c7ea90d@mail.gmail.com> Hi, I tested both packages on my system (Thinkpad T61, Win XP Pro SP3 2GB Intel Core2 duo). The package built by Vince doesn't run -- I have initial splash screen for a short time, and then it exits without any error messages. 01/01 works, I was able to go through the steps of the tutorial without problems. One comment -- when I cut the model nodes from the model hierarchy, the 3d widget still shows the mesh surface, I don't know if this is a known issue. Hope this helps Andriy Fedorov On Sat, Jan 3, 2009 at 11:14 AM, Curtis Lisle wrote: > I have also tested the Win32 version of Slicer3 with IA_FEMesh from > the nightly build from 01/01 on the following systems: > > 1. XP Service Pack 2 32bit on Pentium D with 1Gig RAM - works fine > > 2. Vista Home Basic 32 on Intel Core2 6700 with 4Gig RAM - work fine > > 3. Vista Home Basic running on Parallels over Mac OS X 10.5 on a > powerbook - works except assigning materials on Load/BC menu. I > wasn't able to get the rubber band widget to select nodes for > materials. I am not too worried because of the funny Windows-over- > Parallels-over-Mac configuration. The Mac binary version of > Slicer3-alpha-01-01 works fine, including the node selection widget. > So any Mac users can use the Mac binary during the turorial. > > > Curtis Lisle > curtislisle at knowledgevis.com > > > > On Jan 2, 2009, at 12:12 PM, Vincent A. Magnotta wrote: > >> I have posted a version of Slicer3 version 3.3 that was built under >> Windows to be used for the Meshing tutorial on Wednesday. If anyone >> has >> time, please try to download and test if the version will run on your >> Windows system. The build is missing some standard command line >> modules >> because of compiled size limitations with Visual studio. >> >> http://mri.radiology.uiowa.edu/downloads/Slicer3-3.3-alpha-2008-12-28-win32.zip >> >> Thanks, >> Vince >> >> -- >> Associate Professor >> Department of Radiology >> 0453-D JCP >> 200 Hawkins Drive >> Iowa City, IA 52242 >> E-mail: vincent-magnotta at uiowa.edu >> Phone: 319-356-8255 Fax: 319-353-6275 >> Website: http://mri.radiology.uiowa.edu >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From curtislisle at knowledgevis.com Sat Jan 3 20:37:49 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Sat, 3 Jan 2009 20:37:49 -0500 Subject: [slicer-devel] Windows Slicer 3.3 Alpha Version In-Reply-To: <82f536810901031640l42e900a0j7ae25db71c7ea90d@mail.gmail.com> References: <1230916343.6516.21.camel@ganglia> <92998EB9-CE41-4159-9395-20E19EF00518@knowledgevis.com> <82f536810901031640l42e900a0j7ae25db71c7ea90d@mail.gmail.com> Message-ID: <4B722987-EF4C-42FE-AB13-A7D9A2DCDCFB@knowledgevis.com> We appreciate your testing, Andriy! Yes, this rendering anomaly is a known issue. For the tutorial version this week, the surface and mesh datasets are being rendered separately from their MRML nodes. We have a design in the works to fix this that will be going into the trunk after the AHM. Thanks, again! Curt Curtis Lisle curtislisle at knowledgevis.com On Jan 3, 2009, at 7:40 PM, Andriy Fedorov wrote: > Hi, > > I tested both packages on my system (Thinkpad T61, Win XP Pro SP3 2GB > Intel Core2 duo). > > The package built by Vince doesn't run -- I have initial splash screen > for a short time, and then it exits without any error messages. > > 01/01 works, I was able to go through the steps of the tutorial > without problems. One comment -- when I cut the model nodes from the > model hierarchy, the 3d widget still shows the mesh surface, I don't > know if this is a known issue. > > Hope this helps > > Andriy Fedorov > > > > On Sat, Jan 3, 2009 at 11:14 AM, Curtis Lisle > wrote: >> I have also tested the Win32 version of Slicer3 with IA_FEMesh from >> the nightly build from 01/01 on the following systems: >> >> 1. XP Service Pack 2 32bit on Pentium D with 1Gig RAM - works fine >> >> 2. Vista Home Basic 32 on Intel Core2 6700 with 4Gig RAM - work fine >> >> 3. Vista Home Basic running on Parallels over Mac OS X 10.5 on a >> powerbook - works except assigning materials on Load/BC menu. I >> wasn't able to get the rubber band widget to select nodes for >> materials. I am not too worried because of the funny Windows-over- >> Parallels-over-Mac configuration. The Mac binary version of >> Slicer3-alpha-01-01 works fine, including the node selection widget. >> So any Mac users can use the Mac binary during the turorial. >> >> >> Curtis Lisle >> curtislisle at knowledgevis.com >> >> >> >> On Jan 2, 2009, at 12:12 PM, Vincent A. Magnotta wrote: >> >>> I have posted a version of Slicer3 version 3.3 that was built under >>> Windows to be used for the Meshing tutorial on Wednesday. If anyone >>> has >>> time, please try to download and test if the version will run on >>> your >>> Windows system. The build is missing some standard command line >>> modules >>> because of compiled size limitations with Visual studio. >>> >>> http://mri.radiology.uiowa.edu/downloads/Slicer3-3.3-alpha-2008-12-28-win32.zip >>> >>> Thanks, >>> Vince >>> >>> -- >>> Associate Professor >>> Department of Radiology >>> 0453-D JCP >>> 200 Hawkins Drive >>> Iowa City, IA 52242 >>> E-mail: vincent-magnotta at uiowa.edu >>> Phone: 319-356-8255 Fax: 319-353-6275 >>> Website: http://mri.radiology.uiowa.edu >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> From nagy.attila at yahoo.com Thu Jan 8 16:15:46 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Thu, 8 Jan 2009 13:15:46 -0800 (PST) Subject: [slicer-devel] Packaging question Message-ID: <336855.83745.qm@web37405.mail.mud.yahoo.com> Hi all, Happy new year! :) The problem: last time - back around the end of October - it worked for me. I said getbuildtest.tcl to --pack, and it packed Slicer, and it worked. Now I get this: $ cd Slicer3-3.3-alpha-2009-01-08-solaris8/ $ ls bin include lib share Slicer3 $ ./Slicer3 ld.so.1: Slicer3-real: fatal: libXss.so.1: open failed: No such file or directory Killed child process exited abnormally well... let's try bin/Slicer: $ bin/Slicer3 Error converting runtime path entry "bin/../lib/Teem-1.10" to real path: No such file or directory umm... what? $ ls lib FreeSurfer MGHImageIO SlicerIO GenerateCLP ModuleDescriptionParser tclap GenerateLM MRML TclTk IGT MRMLIDImageIO Teem-1.10.0 InsightToolkit Python vtk-5.2 ITKCommandIO Qdec vtkITK KWWidgets RemoteIO vtkTeem LoadableModule Slicer3 Okay, let's rename teem's dir: $ mv lib/Teem-1.10.0 lib/Teem-1.10 $ bin/Slicer3 ld.so.1: Slicer3-real: fatal: libXss.so.1: open failed: No such file or directory Killed $ ...and I'm stuck here, because libXss.so.1 _is_ there, see: $ find /usr -name libXss.so.1 find: cannot read dir /usr/lib/cc-cfw: Permission denied find: cannot read dir /usr/lib/scn/bin: Permission denied find: cannot read dir /usr/lib/scn/lib: Permission denied find: cannot read dir /usr/sadm/lib/wbem/doc/com: Permission denied find: cannot read dir /usr/sadm/lib/wbem/doc/javax: Permission denied find: cannot read dir /usr/sadm/lib/wbem/doc/resources: Permission denied /usr/X11/lib/amd64/libXss.so.1 /usr/X11/lib/libXss.so.1 $ pkgchk -l -p /usr/X11/lib/libXss.so.1 NOTE: Couldn't lock the package database. Pathname: /usr/X11/lib/libXss.so.1 Type: regular file Expected mode: 0555 Expected owner: root Expected group: bin Expected file size (bytes): 14204 Expected sum(1) of contents: 39788 Expected last modification: nov. 21 22:04:21 2008 Referenced by the following packages: SUNWxwplt Current status: installed (It's the same system where I compiled Slicer, by the way...) I even tried to ln -s this lib into /usr/lib, just to be sure, but no luck... On the other hand, if not packaged and installed, so from the build-dir, Slicer works fine. Please help me, I'm a bit lost now... :( Thanks Attila From nagy.attila at yahoo.com Fri Jan 9 06:14:47 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Fri, 9 Jan 2009 03:14:47 -0800 (PST) Subject: [slicer-devel] [solved] Packaging question Message-ID: <285213.88528.qm@web37407.mail.mud.yahoo.com> Replyig to my own post... :) Softlinking should have worked with libXss.so.1 (since /usr/lib is okay, ie. libs are picked from there), but nope... this is what confused me. Setting the LD_LIBRARY_PATH variable to /usr/X11/lib, and launching Slicer3 works. $ LD_LIBRARY_PATH=/usr/X11/lib; export LD_LIBRARY_PATH; ./Slicer3 Starting Slicer: /build/Desktop/Slicer3.3-alpha/Slicer3-3.3-alpha-2009-01-09-solaris8 Attila --- On Thu, 1/8/09, Attila Nagy wrote: > From: Attila Nagy > Subject: [slicer-devel] Packaging question > To: "Slicer Devel List" > Date: Thursday, January 8, 2009, 10:15 PM > Hi all, > > Happy new year! :) > > The problem: > last time - back around the end of October - it worked for > me. I said getbuildtest.tcl to --pack, and it packed Slicer, > and it worked. > > Now I get this: > > $ cd Slicer3-3.3-alpha-2009-01-08-solaris8/ > $ ls > bin include lib share Slicer3 > $ ./Slicer3 > ld.so.1: Slicer3-real: fatal: libXss.so.1: open failed: No > such file or directory > Killed > > child process exited abnormally > > well... > let's try bin/Slicer: > > $ bin/Slicer3 > Error converting runtime path entry > "bin/../lib/Teem-1.10" to real path: No such file > or directory > > umm... what? > > $ ls lib > FreeSurfer MGHImageIO SlicerIO > GenerateCLP ModuleDescriptionParser tclap > GenerateLM MRML TclTk > IGT MRMLIDImageIO > Teem-1.10.0 > InsightToolkit Python vtk-5.2 > ITKCommandIO Qdec vtkITK > KWWidgets RemoteIO vtkTeem > LoadableModule Slicer3 > > Okay, let's rename teem's dir: > $ mv lib/Teem-1.10.0 lib/Teem-1.10 > $ bin/Slicer3 > ld.so.1: Slicer3-real: fatal: libXss.so.1: open failed: No > such file or directory > Killed > $ > > ...and I'm stuck here, because libXss.so.1 _is_ there, > see: > > $ find /usr -name libXss.so.1 > find: cannot read dir /usr/lib/cc-cfw: Permission denied > find: cannot read dir /usr/lib/scn/bin: Permission denied > find: cannot read dir /usr/lib/scn/lib: Permission denied > find: cannot read dir /usr/sadm/lib/wbem/doc/com: > Permission denied > find: cannot read dir /usr/sadm/lib/wbem/doc/javax: > Permission denied > find: cannot read dir /usr/sadm/lib/wbem/doc/resources: > Permission denied > /usr/X11/lib/amd64/libXss.so.1 > /usr/X11/lib/libXss.so.1 > $ pkgchk -l -p /usr/X11/lib/libXss.so.1 > NOTE: Couldn't lock the package database. > Pathname: /usr/X11/lib/libXss.so.1 > Type: regular file > Expected mode: 0555 > Expected owner: root > Expected group: bin > Expected file size (bytes): 14204 > Expected sum(1) of contents: 39788 > Expected last modification: nov. 21 22:04:21 2008 > Referenced by the following packages: > SUNWxwplt > Current status: installed > > (It's the same system where I compiled Slicer, by the > way...) > I even tried to ln -s this lib into /usr/lib, just to be > sure, but no luck... > > On the other hand, if not packaged and installed, so from > the build-dir, Slicer works fine. > > Please help me, I'm a bit lost now... :( > > Thanks > > Attila > > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with > unsubscribe as the subject From kikinis at bwh.harvard.edu Sat Jan 10 00:46:24 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Sat, 10 Jan 2009 00:46:24 -0500 Subject: [slicer-devel] Thank you for SLC and Slicer 3.4 preparations Message-ID: <49683630.3040903@bwh.harvard.edu> http://wiki.slicer.org/slicerWiki/index.php/Slicer3:Events Hello Slicer Developers, I would like to thank all of you for making the Project Week and AHM a great success. A special thank you to Deb Zemek and Katie Mastrogiacomo who have ably organized the meeting with guidance from Tina Kapur. As announced at the project week, we will initiate a new release of Slicer, Slicer 3.4. Please see the link above for more information. For your convenience, I have pasted the text below. all the best to all of you and see you in June in Boston Ron Slicer3.4 A product release of Slicer 3.4 is scheduled for Feb/March * There will be a code freeze on Feb 4, 2009 * Requirements for modules to be added: o The module is feature complete for the tasks advertised o The module has a test. See here for more information. o Module has documentation on the Slicer wiki. Please use the template provided here to structure your page. Please add a pointer to the documentation on the Slicer wiki to the the Help tab of the module. See the editor module in Slicer for an example. o The contributor (and their manager/advisor, lab and funding source) are listed in the Acknowledgement tab of the module. Please see the Models module for and example. The people listed in the acknowledgement will be the primary people for support and maintenance relative of the module. * After the freeze date, we will only accept bug fixes but no new features. * If your module has documentation in Slicer 3.2, please copy/paste/update into the 3.4 version * Please contact Ron Kikinis, if you had a module in 3.2 and you do not want it to be in 3.4. -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From vrnova at gmail.com Mon Jan 12 15:16:28 2009 From: vrnova at gmail.com (Yanling Liu) Date: Mon, 12 Jan 2009 15:16:28 -0500 Subject: [slicer-devel] register images module Message-ID: <24ceacd90901121216t338345b1q54f0a2e39f2a692e@mail.gmail.com> Hello, I am wondering why the register images module is being comment out in the Applications/CLI/CMakeLists.txt. Is this module safe to use? Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090112/e4bf9a82/attachment-0001.html From pieper at bwh.harvard.edu Mon Jan 12 17:18:36 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 12 Jan 2009 17:18:36 -0500 Subject: [slicer-devel] register images module In-Reply-To: <24ceacd90901121216t338345b1q54f0a2e39f2a692e@mail.gmail.com> References: <24ceacd90901121216t338345b1q54f0a2e39f2a692e@mail.gmail.com> Message-ID: <496BC1BC.2030900@bwh.harvard.edu> Hi Yanling - There were some build problems. Katie and Stephen are working on them. If it builds on your machine you should go ahead and use it. -Steve Yanling Liu wrote: > Hello, > > I am wondering why the register images module is being comment out in > the Applications/CLI/CMakeLists.txt. Is this module safe to use? > > Thanks, > Yanling > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From carlos.sanchez.mendoza at gmail.com Tue Jan 13 05:24:32 2009 From: carlos.sanchez.mendoza at gmail.com (=?UTF-8?Q?Carlos_S=C3=A1nchez_Mendoza?=) Date: Tue, 13 Jan 2009 11:24:32 +0100 Subject: [slicer-devel] Modules as shared libraries (Carlos S. Mendoza) Message-ID: Hi all, I am trying to improve performance in my modules by deploying them as shared libraries instead of stand-alone executables, avoiding the file system. I replaced add_executable(MyFilter ${MyFilter_SOURCE}) target_link_libraries(MyFilter ${ITK_LIBRARIES}) slicer3_set_plugins_output_path(MyFilter) with add_library(MyFilterLib SHARED ${MyFilter_SOURCE}) set_target_properties(MyFilterLib PROPERTIES COMPILE_FLAGS "-Dmain=ModuleEntryPoint") target_link_libraries(MyFilterLib ${ITK_LIBRARIES}) slicer3_set_plugins_output_path(MyFilterLib) After CMaking and compiling, I made Slicer detect the new module but nothing happens when I click on Apply (the time display stays in 0.00 s) What step am I missing? Thank you in advance. Carlos S. Mendoza. University of Sevilla, Spain. From curtislisle at knowledgevis.com Tue Jan 13 10:09:04 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Tue, 13 Jan 2009 10:09:04 -0500 Subject: [slicer-devel] Modules as shared libraries (Carlos S. Mendoza) In-Reply-To: References: Message-ID: <19CAACCB-D418-4BDB-AB8B-D243BF517D76@knowledgevis.com> Hi Carlos, It sounds like you are converting from a command line module into a loadable module. It may be that your library call from KWWidgets is not being resolved at run-time. Loadable modules can be added to Slicer pretty easily, but you may be missing a call in the CMakeLists.txt which lets Slicer discover the libraries automatically. Please look at the CMakeLists file from the module GradientAnisotropicDiffusionFilter as an example of the few, key macros needed for loadable modules. In particular, you may be missing the "generate_lm()" macro, which makes sure the shared library is accessible dynamically and the slicer3_install_modules() macro, which copies it into a standard location for loadable modules. Good luck! Curtis Curtis Lisle curtislisle at knowledgevis.com On Jan 13, 2009, at 5:24 AM, Carlos S?nchez Mendoza wrote: > Hi all, > > I am trying to improve performance in my modules by deploying them as > shared libraries instead of stand-alone executables, avoiding the file > system. > > I replaced > > add_executable(MyFilter ${MyFilter_SOURCE}) > target_link_libraries(MyFilter ${ITK_LIBRARIES}) > slicer3_set_plugins_output_path(MyFilter) > > with > > add_library(MyFilterLib SHARED ${MyFilter_SOURCE}) > set_target_properties(MyFilterLib PROPERTIES COMPILE_FLAGS > "-Dmain=ModuleEntryPoint") > target_link_libraries(MyFilterLib ${ITK_LIBRARIES}) > slicer3_set_plugins_output_path(MyFilterLib) > > After CMaking and compiling, I made Slicer detect the new module but > nothing happens when I click on Apply (the time display stays in 0.00 > s) > > What step am I missing? > > Thank you in advance. > > Carlos S. Mendoza. > University of Sevilla, Spain. > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From carlos.sanchez.mendoza at gmail.com Tue Jan 13 11:00:52 2009 From: carlos.sanchez.mendoza at gmail.com (=?UTF-8?Q?Carlos_S=C3=A1nchez_Mendoza?=) Date: Tue, 13 Jan 2009 17:00:52 +0100 Subject: [slicer-devel] Modules as shared libraries (Carlos S. Mendoza) In-Reply-To: <19CAACCB-D418-4BDB-AB8B-D243BF517D76@knowledgevis.com> References: <19CAACCB-D418-4BDB-AB8B-D243BF517D76@knowledgevis.com> Message-ID: Curtis, sorry for the burden, I just feel so close to the solution... Your hint didn't solve the problem. Let me describe a little what I intend to do. I have a command-line module consisting of just a main method and a GenerateCLP header. Steve Pieper told me once (during my visit last summer) I just had to change a couple lines in the CMakeLists file to make it communicate directly to the MRML scene. I embedded below the CMakeLists I am trying to use. Is that enough? Do I need to change main() for ModuleEntryPoint() as suggested in the Execution Model? Thanks so much. I am addressing this to the community in case someone else has time for this. Carlos S. Mendoza University of Sevilla, Spain. project(MultiSeedAutomaticARGlib) cmake_minimum_required(VERSION 2.6) find_package(Slicer3 REQUIRED) include(${Slicer3_USE_FILE}) slicer3_set_default_install_prefix_for_external_projects() FIND_PACKAGE(ITK REQUIRED) IF(ITK_FOUND) INCLUDE(${ITK_USE_FILE}) ENDIF(ITK_FOUND) set(MultiSeedAutomaticARGlib_SOURCE MultiSeedAutomaticARGlib.cxx ) GENERATECLP(MultiSeedAutomaticARGlib_SOURCE MultiSeedAutomaticARGlib.xml) GENERATELM(MultiSeedAutomaticARGlib_Source MultiSeedAutomaticARGlib.xml) add_library(MultiSeedAutomaticARGlib SHARED ${MultiSeedAutomaticARGlib_SOURCE}) set_target_properties(MultiSeedAutomaticARGlib PROPERTIES COMPILE_FLAGS "-Dmain=ModuleEntryPoint") target_link_libraries(MultiSeedAutomaticARGlib ITKIO ITKCommon) slicer3_set_plugins_output_path(MultiSeedAutomaticARGlib) slicer3_install_modules() 2009/1/13 Curtis Lisle : > Hi Carlos, > It sounds like you are converting from a command line module into a > loadable module. It may be that your library call from KWWidgets is not > being resolved at run-time. Loadable modules can be added to Slicer pretty > easily, but you may be missing a call in the CMakeLists.txt which lets > Slicer discover the libraries automatically. > > Please look at the CMakeLists file from the module > GradientAnisotropicDiffusionFilter as an example of the few, key macros > needed for loadable modules. In particular, you may be missing the > "generate_lm()" macro, which makes sure the shared library is accessible > dynamically and the slicer3_install_modules() macro, which copies it into a > standard location for loadable modules. Good luck! > > Curtis > > Curtis Lisle > curtislisle at knowledgevis.com > > > > On Jan 13, 2009, at 5:24 AM, Carlos S?nchez Mendoza wrote: > >> Hi all, >> >> I am trying to improve performance in my modules by deploying them as >> shared libraries instead of stand-alone executables, avoiding the file >> system. >> >> I replaced >> >> add_executable(MyFilter ${MyFilter_SOURCE}) >> target_link_libraries(MyFilter ${ITK_LIBRARIES}) >> slicer3_set_plugins_output_path(MyFilter) >> >> with >> >> add_library(MyFilterLib SHARED ${MyFilter_SOURCE}) >> set_target_properties(MyFilterLib PROPERTIES COMPILE_FLAGS >> "-Dmain=ModuleEntryPoint") >> target_link_libraries(MyFilterLib ${ITK_LIBRARIES}) >> slicer3_set_plugins_output_path(MyFilterLib) >> >> After CMaking and compiling, I made Slicer detect the new module but >> nothing happens when I click on Apply (the time display stays in 0.00 >> s) >> >> What step am I missing? >> >> Thank you in advance. >> >> Carlos S. Mendoza. >> University of Sevilla, Spain. >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > > From carlos.sanchez.mendoza at gmail.com Tue Jan 13 13:32:45 2009 From: carlos.sanchez.mendoza at gmail.com (=?UTF-8?Q?Carlos_S=C3=A1nchez_Mendoza?=) Date: Tue, 13 Jan 2009 19:32:45 +0100 Subject: [slicer-devel] Modules as shared libraries (Carlos S. Mendoza) In-Reply-To: References: <19CAACCB-D418-4BDB-AB8B-D243BF517D76@knowledgevis.com> Message-ID: Hi all, strangely enough I just realized that the module is actually executing behind the scenes even though the timer stays in 0.00 seconds. After some time the output is displayed. It would be very nice to see the timer change, so that I can actually check if the shared object starts right away (proving that the file system is not being used). Any ideas on why the timer doesn't work when turning a command-line module into a shared object? Thanks Carlos S. Mendoza University of Sevilla, Spain. 2009/1/13 Carlos S?nchez Mendoza : > Curtis, > > sorry for the burden, I just feel so close to the solution... Your > hint didn't solve the problem. > > Let me describe a little what I intend to do. > > I have a command-line module consisting of just a main method and a > GenerateCLP header. Steve Pieper told me once (during my visit last > summer) I just had to change a couple lines in the CMakeLists file to > make it communicate directly to the MRML scene. > > I embedded below the CMakeLists I am trying to use. Is that enough? Do > I need to change main() for ModuleEntryPoint() as suggested in the > Execution Model? > > Thanks so much. I am addressing this to the community in case someone > else has time for this. > > Carlos S. Mendoza > University of Sevilla, Spain. > > > > project(MultiSeedAutomaticARGlib) > > cmake_minimum_required(VERSION 2.6) > > find_package(Slicer3 REQUIRED) > include(${Slicer3_USE_FILE}) > > slicer3_set_default_install_prefix_for_external_projects() > > FIND_PACKAGE(ITK REQUIRED) > IF(ITK_FOUND) > INCLUDE(${ITK_USE_FILE}) > ENDIF(ITK_FOUND) > > set(MultiSeedAutomaticARGlib_SOURCE MultiSeedAutomaticARGlib.cxx ) > > GENERATECLP(MultiSeedAutomaticARGlib_SOURCE MultiSeedAutomaticARGlib.xml) > > GENERATELM(MultiSeedAutomaticARGlib_Source MultiSeedAutomaticARGlib.xml) > > add_library(MultiSeedAutomaticARGlib SHARED ${MultiSeedAutomaticARGlib_SOURCE}) > set_target_properties(MultiSeedAutomaticARGlib PROPERTIES > COMPILE_FLAGS "-Dmain=ModuleEntryPoint") > target_link_libraries(MultiSeedAutomaticARGlib ITKIO ITKCommon) > slicer3_set_plugins_output_path(MultiSeedAutomaticARGlib) > > slicer3_install_modules() > > > > > 2009/1/13 Curtis Lisle : >> Hi Carlos, >> It sounds like you are converting from a command line module into a >> loadable module. It may be that your library call from KWWidgets is not >> being resolved at run-time. Loadable modules can be added to Slicer pretty >> easily, but you may be missing a call in the CMakeLists.txt which lets >> Slicer discover the libraries automatically. >> >> Please look at the CMakeLists file from the module >> GradientAnisotropicDiffusionFilter as an example of the few, key macros >> needed for loadable modules. In particular, you may be missing the >> "generate_lm()" macro, which makes sure the shared library is accessible >> dynamically and the slicer3_install_modules() macro, which copies it into a >> standard location for loadable modules. Good luck! >> >> Curtis >> >> Curtis Lisle >> curtislisle at knowledgevis.com >> >> >> >> On Jan 13, 2009, at 5:24 AM, Carlos S?nchez Mendoza wrote: >> >>> Hi all, >>> >>> I am trying to improve performance in my modules by deploying them as >>> shared libraries instead of stand-alone executables, avoiding the file >>> system. >>> >>> I replaced >>> >>> add_executable(MyFilter ${MyFilter_SOURCE}) >>> target_link_libraries(MyFilter ${ITK_LIBRARIES}) >>> slicer3_set_plugins_output_path(MyFilter) >>> >>> with >>> >>> add_library(MyFilterLib SHARED ${MyFilter_SOURCE}) >>> set_target_properties(MyFilterLib PROPERTIES COMPILE_FLAGS >>> "-Dmain=ModuleEntryPoint") >>> target_link_libraries(MyFilterLib ${ITK_LIBRARIES}) >>> slicer3_set_plugins_output_path(MyFilterLib) >>> >>> After CMaking and compiling, I made Slicer detect the new module but >>> nothing happens when I click on Apply (the time display stays in 0.00 >>> s) >>> >>> What step am I missing? >>> >>> Thank you in advance. >>> >>> Carlos S. Mendoza. >>> University of Sevilla, Spain. >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >> >> > From curtislisle at knowledgevis.com Tue Jan 13 15:10:50 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Tue, 13 Jan 2009 15:10:50 -0500 Subject: [slicer-devel] Modules as shared libraries (Carlos S. Mendoza) In-Reply-To: References: <19CAACCB-D418-4BDB-AB8B-D243BF517D76@knowledgevis.com> Message-ID: Unfortunately I haven't come across this before. The command line modules I've written didn't update status on the GUI. From the previous discussion, I remain a bit concerned that your module might have trouble packaging into Slicer later because of a CMake process that includes macros for both a command line and a shared module. Most work I'm familiar with has chosen one or the other method of integration. At some point, you may want to convert all the way over to a loadable module, which would only take a few hours of editing, based on the GAD filter example. I hope this helps. Take care! Curtis Curtis Lisle curtislisle at knowledgevis.com On Jan 13, 2009, at 1:32 PM, Carlos S?nchez Mendoza wrote: > Hi all, > > strangely enough I just realized that the module is actually executing > behind the scenes even though the timer stays in 0.00 seconds. After > some time the output is displayed. It would be very nice to see the > timer change, so that I can actually check if the shared object starts > right away (proving that the file system is not being used). > > Any ideas on why the timer doesn't work when turning a command-line > module into a shared object? > > Thanks > > Carlos S. Mendoza > University of Sevilla, Spain. > > 2009/1/13 Carlos S?nchez Mendoza : >> Curtis, >> >> sorry for the burden, I just feel so close to the solution... Your >> hint didn't solve the problem. >> >> Let me describe a little what I intend to do. >> >> I have a command-line module consisting of just a main method and a >> GenerateCLP header. Steve Pieper told me once (during my visit last >> summer) I just had to change a couple lines in the CMakeLists file to >> make it communicate directly to the MRML scene. >> >> I embedded below the CMakeLists I am trying to use. Is that enough? >> Do >> I need to change main() for ModuleEntryPoint() as suggested in the >> Execution Model? >> >> Thanks so much. I am addressing this to the community in case someone >> else has time for this. >> >> Carlos S. Mendoza >> University of Sevilla, Spain. >> >> >> >> project(MultiSeedAutomaticARGlib) >> >> cmake_minimum_required(VERSION 2.6) >> >> find_package(Slicer3 REQUIRED) >> include(${Slicer3_USE_FILE}) >> >> slicer3_set_default_install_prefix_for_external_projects() >> >> FIND_PACKAGE(ITK REQUIRED) >> IF(ITK_FOUND) >> INCLUDE(${ITK_USE_FILE}) >> ENDIF(ITK_FOUND) >> >> set(MultiSeedAutomaticARGlib_SOURCE MultiSeedAutomaticARGlib.cxx ) >> >> GENERATECLP(MultiSeedAutomaticARGlib_SOURCE >> MultiSeedAutomaticARGlib.xml) >> >> GENERATELM(MultiSeedAutomaticARGlib_Source >> MultiSeedAutomaticARGlib.xml) >> >> add_library(MultiSeedAutomaticARGlib SHARED $ >> {MultiSeedAutomaticARGlib_SOURCE}) >> set_target_properties(MultiSeedAutomaticARGlib PROPERTIES >> COMPILE_FLAGS "-Dmain=ModuleEntryPoint") >> target_link_libraries(MultiSeedAutomaticARGlib ITKIO ITKCommon) >> slicer3_set_plugins_output_path(MultiSeedAutomaticARGlib) >> >> slicer3_install_modules() >> >> >> >> >> 2009/1/13 Curtis Lisle : >>> Hi Carlos, >>> It sounds like you are converting from a command line module into a >>> loadable module. It may be that your library call from KWWidgets >>> is not >>> being resolved at run-time. Loadable modules can be added to >>> Slicer pretty >>> easily, but you may be missing a call in the CMakeLists.txt which >>> lets >>> Slicer discover the libraries automatically. >>> >>> Please look at the CMakeLists file from the module >>> GradientAnisotropicDiffusionFilter as an example of the few, key >>> macros >>> needed for loadable modules. In particular, you may be missing the >>> "generate_lm()" macro, which makes sure the shared library is >>> accessible >>> dynamically and the slicer3_install_modules() macro, which copies >>> it into a >>> standard location for loadable modules. Good luck! >>> >>> Curtis >>> >>> Curtis Lisle >>> curtislisle at knowledgevis.com >>> >>> >>> >>> On Jan 13, 2009, at 5:24 AM, Carlos S?nchez Mendoza wrote: >>> >>>> Hi all, >>>> >>>> I am trying to improve performance in my modules by deploying >>>> them as >>>> shared libraries instead of stand-alone executables, avoiding the >>>> file >>>> system. >>>> >>>> I replaced >>>> >>>> add_executable(MyFilter ${MyFilter_SOURCE}) >>>> target_link_libraries(MyFilter ${ITK_LIBRARIES}) >>>> slicer3_set_plugins_output_path(MyFilter) >>>> >>>> with >>>> >>>> add_library(MyFilterLib SHARED ${MyFilter_SOURCE}) >>>> set_target_properties(MyFilterLib PROPERTIES COMPILE_FLAGS >>>> "-Dmain=ModuleEntryPoint") >>>> target_link_libraries(MyFilterLib ${ITK_LIBRARIES}) >>>> slicer3_set_plugins_output_path(MyFilterLib) >>>> >>>> After CMaking and compiling, I made Slicer detect the new module >>>> but >>>> nothing happens when I click on Apply (the time display stays in >>>> 0.00 >>>> s) >>>> >>>> What step am I missing? >>>> >>>> Thank you in advance. >>>> >>>> Carlos S. Mendoza. >>>> University of Sevilla, Spain. >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as >>>> the >>>> subject >>> >>> >> From fedorov at bwh.harvard.edu Wed Jan 14 08:35:02 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Wed, 14 Jan 2009 08:35:02 -0500 Subject: [slicer-devel] Fwd: [Insight-users] Nokia Qt Adopts LGPL Message-ID: <82f536810901140535k18d1fa1eo608886a7fe66621d@mail.gmail.com> Date: Wed, 14 Jan 2009 09:47:54 +0100 From: Iv?n Mac?a Subject: [Insight-users] Nokia Qt Adopts LGPL To: "Insight Developers" , "Insight Users" Dear all, I think this is very interesting for the ITK community, specially for those involved in commercial developments. http://arstechnica.com/news.ars/post/20090114-nokia-qt-lgpl-switch-huge-win- for-cross-platform-development.html Best regards Iv?n From fedorov at bwh.harvard.edu Wed Jan 14 20:05:19 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Wed, 14 Jan 2009 20:05:19 -0500 Subject: [slicer-devel] Testing wizard-based module Message-ID: <82f536810901141705o1fa4335fxb0a67f460f05800b@mail.gmail.com> Hi, I have a command line application for my wizard-based module, that will invoke the processing steps from my module logic. I am in the process of preparing the tests for the module's functionality. I have two general questions: 1) Since my module consists of multiple steps, I would like to have a test for each step separately, and then a test that includes the complete workflow too. What is the proper way to do this: to have separate binaries for each step/level, or have extra options in the xml description to instruct command line module to exit after a specified processing step? 2) What is the right way to do regression testing for rigid registration (one of the steps in my module)? If I save the transform, it may be different up to some degree on another platform due to different architecture, and the difference may not mean that the test failed. Should I compare registered images then instead of the transform? Thanks Andriy Fedorov From bill.lorensen at gmail.com Wed Jan 14 23:07:41 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Wed, 14 Jan 2009 23:07:41 -0500 Subject: [slicer-devel] Testing wizard-based module In-Reply-To: <82f536810901141705o1fa4335fxb0a67f460f05800b@mail.gmail.com> References: <82f536810901141705o1fa4335fxb0a67f460f05800b@mail.gmail.com> Message-ID: <4db4735c0901142007v3aa4b939ib96c6c18e77f97e6@mail.gmail.com> For the test, I would use one executable with multiple options. The same executable can be used in multiple tests. For the regression test, I would compare the resulting images. If there are slight differences, they can be accommodated by the testing framework used in Applications/CLI/Testing. Bill On Wed, Jan 14, 2009 at 8:05 PM, Andriy Fedorov wrote: > Hi, > > I have a command line application for my wizard-based module, that > will invoke the processing steps from my module logic. I am in the > process of preparing the tests for the module's functionality. I have > two general questions: > > 1) Since my module consists of multiple steps, I would like to have a > test for each step separately, and then a test that includes the > complete workflow too. What is the proper way to do this: to have > separate binaries for each step/level, or have extra options in the > xml description to instruct command line module to exit after a > specified processing step? > > 2) What is the right way to do regression testing for rigid > registration (one of the steps in my module)? If I save the transform, > it may be different up to some degree on another platform due to > different architecture, and the difference may not mean that the test > failed. Should I compare registered images then instead of the > transform? > > Thanks > > Andriy Fedorov > From kikinis at bwh.harvard.edu Thu Jan 15 09:36:28 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis MD) Date: Thu, 15 Jan 2009 09:36:28 -0500 Subject: [slicer-devel] [Fwd: [Active-system-users] Web services need update] Message-ID: <496F49EC.80906@bwh.harvard.edu> Hi, this will likely affect the slicer svn and slicer mailing lists. Ron -------- Original Message -------- Subject: [Active-system-users] Web services need update Date: Mon, 12 Jan 2009 10:30:51 -0500 From: Bob McKie To: active-system-users at bwh.harvard.edu We have been informed by Partners Inforrmation Security that our Apache webserver has security vulnerabilities that we must address in order to continue offering outside access to our many web services.. "help" built a new version of Apache and plan to install this on Thursday, January 15, starting at 11:00 AM. This means that ALL webservices will be unavailable from 11:00 AM until we successfully repair the webserver. We are hoping that this will be a 1-hour repair, but if our handbuilt version does NOT work properly, we'll have to revert to the old version, which will take some time. We are hoping that this can be completed by 1:00 PM, but once we start, we have to work until completion, regardless of how long it takes to complete.. If anyone reading this has a serious objection to this time slot for the proposed repair, respond to "help" ASAP. We're sorry for the inconvenience, but Infosec has the upper hand in this matter. Thanks, Bob McKie -- Robert A. McKie Director of Information Technology Brigham and Women's Hospital Surgical Planning Lab _______________________________________________ Active-system-users mailing list Active-system-users at bwh.harvard.edu http://massmail.spl.harvard.edu/mailman/listinfo/active-system-users -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From kikinis at bwh.harvard.edu Thu Jan 15 12:53:01 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Thu, 15 Jan 2009 12:53:01 -0500 Subject: [slicer-devel] [Fwd: [Active-system-users] Webserver upgrade was unsuccessful] Message-ID: <496F77FD.9070201@bwh.harvard.edu> -------- Original Message -------- Subject: [Active-system-users] Webserver upgrade was unsuccessful Date: Thu, 15 Jan 2009 12:32:39 -0500 From: Bob McKie To: active-system-users at bwh.harvard.edu, namic-all at na-mic.org, "zack.galbreath at kitware.com >> Zack Galbreath" We ran into a snag which will take some time to diagnose. We decided to test our new apache install on the machine hosting the publication database first, which failed, so we restored apache from tape backup. The Publications Database seems to be working OK - let us know if you find otherwise. Further interruptions of web services have been postponed until a later date. Everything should be working as it was early this morning. Thanks, Bob McKie -- Robert A. McKie Director of Information Technology Brigham and Women's Hospital Surgical Planning Lab _______________________________________________ Active-system-users mailing list Active-system-users at bwh.harvard.edu http://massmail.spl.harvard.edu/mailman/listinfo/active-system-users -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From nagy.attila at yahoo.com Mon Jan 19 06:44:58 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Mon, 19 Jan 2009 03:44:58 -0800 (PST) Subject: [slicer-devel] cmcurl source "cleaning" Message-ID: <391516.79091.qm@web37404.mail.mud.yahoo.com> Dear all, please could somebody clean the cmcurl source, because it has ^M line endings in quite a bunch of files, and it breaks the compilaion with Sun's compiler: www.orl.szote.u-szeged.hu/~attila/cmcurl.build.suncc.txt all these errors go away, if I do a dos2unix conversion on the whole cmcurl subdirectory. I think, it shouldn't break other compilers/setups. Thanks, Attila From pieper at bwh.harvard.edu Mon Jan 19 11:36:05 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 19 Jan 2009 11:36:05 -0500 Subject: [slicer-devel] cmcurl source "cleaning" In-Reply-To: <391516.79091.qm@web37404.mail.mud.yahoo.com> References: <391516.79091.qm@web37404.mail.mud.yahoo.com> Message-ID: <4974ABF5.3090509@bwh.harvard.edu> Hi Attila - Done... http://viewvc.slicer.org/viewcvs.cgi?rev=126&root=Slicer3-lib-mirrors&view=rev Thanks for the report, -Steve Attila Nagy wrote: > Dear all, > > please could somebody clean the cmcurl source, because it has ^M line endings in quite a bunch of files, and it breaks the compilaion with Sun's compiler: > www.orl.szote.u-szeged.hu/~attila/cmcurl.build.suncc.txt > > all these errors go away, if I do a dos2unix conversion on the whole cmcurl subdirectory. > > I think, it shouldn't break other compilers/setups. > > Thanks, > > Attila > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From nagy.attila at yahoo.com Mon Jan 19 11:41:21 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Mon, 19 Jan 2009 08:41:21 -0800 (PST) Subject: [slicer-devel] cmcurl source "cleaning" In-Reply-To: <4974ABF5.3090509@bwh.harvard.edu> Message-ID: <268258.58048.qm@web37407.mail.mud.yahoo.com> Cool, thanks! :) --- On Mon, 1/19/09, Steve Pieper wrote: > From: Steve Pieper > Subject: Re: [slicer-devel] cmcurl source "cleaning" > To: nagy.attila at yahoo.com > Cc: "Slicer Devel List" > Date: Monday, January 19, 2009, 5:36 PM > Hi Attila - > > Done... > > http://viewvc.slicer.org/viewcvs.cgi?rev=126&root=Slicer3-lib-mirrors&view=rev > > Thanks for the report, > -Steve > > Attila Nagy wrote: > > Dear all, > > > > please could somebody clean the cmcurl source, because > it has ^M line endings in quite a bunch of files, and it > breaks the compilaion with Sun's compiler: > > > www.orl.szote.u-szeged.hu/~attila/cmcurl.build.suncc.txt > > > > all these errors go away, if I do a dos2unix > conversion on the whole cmcurl subdirectory. > > > > I think, it shouldn't break other > compilers/setups. > > > > Thanks, > > > > Attila > > > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with > unsubscribe as the subject From fedorov at bwh.harvard.edu Mon Jan 19 15:26:50 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 19 Jan 2009 15:26:50 -0500 Subject: [slicer-devel] Rigid registration reproducibility Message-ID: <82f536810901191226j63ffd262w3787cc9149f97209@mail.gmail.com> Hi, I observe that Rigid registration CLI module produces different (difference in the order of 0.1 for translation) transforms when run through GUI and using the command line interface. I run these on the same machine, and the metric is initialized with the same seed. The transform is consistent when I repeatedly run it from command line or GUI, but not between command line and GUI results. I would expect the same result The test images I use are available in the trunk: $Slicer3/Testing/Data/Input/MRMeningioma0.nrrd (fixed) and $Slicer3/Testing/Data/Input/MRMeningioma1.nrrd (moving). Slicer GUI invocation: slicer:0x7f8604605652 --processinformationaddress 0x4734f28 --fixedsmoothingfactor 0 --movingsmoothingfactor 0 --histogrambins 30 --spatialsamples 10000 --iterations 100,100,50,20 --learningrate 0.01,0.005,0.0005,0.0002 --translationscale 100 --outputtransform /tmp/Slicer3andrey/CHDDI_vtkMRMLLinearTransformNodeE.txt slicer:0x104e0f0#vtkMRMLScalarVolumeNode1 slicer:0x104e0f0#vtkMRMLScalarVolumeNode1 Transform: Baseline transform: QuaternionRigidTransform (0x6dd7c0) RTTI typeinfo: itk::QuaternionRigidTransform Reference Count: 4 Modified Time: 230 Debug: Off Observers: none Matrix: 0.999645 -0.0259667 0.00608542 0.0260626 0.999529 -0.0162505 -0.00566058 0.0164033 0.99985 Offset: [7.54937, 5.35469, -0.0257966] Center: [0.00025, 0.00025, 0] Translation: [7.54936, 5.3547, -0.0257939] Inverse: 0.999644 0.0260626 -0.00566058 -0.0259666 0.999528 0.0164033 0.00608542 -0.0162505 0.999849 Singular: 0 Rotation: 0.00816444 0.00293686 0.0130089 0.999878 Transform matrix: 0.999895 -0.0132102 0.0059665 0.0132875 0.999826 -0.0130963 -0.00579246 0.0131742 0.999896 Slicer command line invocation: ${Slicer3_HOME}/Slicer3 --launch RigidRegistration --iterations 100,100,50,20 --fixedsmoothingfactor 0 --movingsmoothingfactor 0 --histogrambins 30 --spatialsamples 10000 --learningrate 0.01,0.005,0.0005,0.0002 --translationscale 100 --resampledmovingfilename /home/andrey/local/src/Slicer/Slicer3-build/Testing/Temporary/RigidRegistrationRegressionTestOutput.nrrd --outputtransform /home/andrey/local/src/Slicer/Slicer3-build/Testing/Temporary/RigidRegistrationRegressionTestTransform.txt /home/andrey/local/src/Slicer/Slicer3/Testing/Data/Input/MRMeningioma0.nrrd /home/andrey/local/src/Slicer/Slicer3/Testing/Data/Input/MRMeningioma1.nrrd Transform: Output transform: QuaternionRigidTransform (0x6dda20) RTTI typeinfo: itk::QuaternionRigidTransform Reference Count: 4 Modified Time: 235 Debug: Off Observers: none Matrix: 0.999895 -0.0132102 0.0059665 0.0132875 0.999826 -0.0130963 -0.00579246 0.0131742 0.999896 Offset: [7.76758, 5.81825, 0.164242] Center: [0.00025, 0.00025, -1.42109e-14] Translation: [7.76758, 5.81825, 0.164244] Inverse: 0.999895 0.0132875 -0.00579246 -0.0132103 0.999826 0.0131742 0.0059665 -0.0130963 0.999897 Singular: 0 Rotation: 0.00656795 0.00293988 0.00662475 0.999952 -- Andriy Fedorov From pieper at bwh.harvard.edu Thu Jan 22 16:49:00 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 22 Jan 2009 16:49:00 -0500 Subject: [slicer-devel] [Fwd: Thank you for SLC and Slicer 3.4 preparations] Message-ID: <4978E9CC.1040403@bwh.harvard.edu> Reminder: We're planning to freeze the slicer3 svn trunk on Feb 4th. This means developers should be hard at work finalizing any code they want to have included in the 3.4 release builds. Be sure to include the documentation and other info Ron outlined below. During February, the nightly builds will be called 3.4 beta and we will be working on bug fixes and performance improvements. By March 1 (if not sooner) we will create a 3.4 release branch, at which point we'll reopen the trunk as 3.5 alpha. Thank you all for your contributions and let us know if there are any questions. -Steve -------- Original Message -------- Subject: [slicer-devel] Thank you for SLC and Slicer 3.4 preparations Date: Sat, 10 Jan 2009 00:46:24 -0500 From: Ron Kikinis To: slicer-devel at bwh.harvard.edu , namic-all at na-mic.org http://wiki.slicer.org/slicerWiki/index.php/Slicer3:Events Hello Slicer Developers, I would like to thank all of you for making the Project Week and AHM a great success. A special thank you to Deb Zemek and Katie Mastrogiacomo who have ably organized the meeting with guidance from Tina Kapur. As announced at the project week, we will initiate a new release of Slicer, Slicer 3.4. Please see the link above for more information. For your convenience, I have pasted the text below. all the best to all of you and see you in June in Boston Ron Slicer3.4 A product release of Slicer 3.4 is scheduled for Feb/March * There will be a code freeze on Feb 4, 2009 * Requirements for modules to be added: o The module is feature complete for the tasks advertised o The module has a test. See here for more information. o Module has documentation on the Slicer wiki. Please use the template provided here to structure your page. Please add a pointer to the documentation on the Slicer wiki to the the Help tab of the module. See the editor module in Slicer for an example. o The contributor (and their manager/advisor, lab and funding source) are listed in the Acknowledgement tab of the module. Please see the Models module for and example. The people listed in the acknowledgement will be the primary people for support and maintenance relative of the module. * After the freeze date, we will only accept bug fixes but no new features. * If your module has documentation in Slicer 3.2, please copy/paste/update into the 3.4 version * Please contact Ron Kikinis, if you had a module in 3.2 and you do not want it to be in 3.4. -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis _______________________________________________ slicer-devel mailing list slicer-devel at bwh.harvard.edu http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From Abdelrahman.Elhaddad at bibalex.org Sun Jan 25 09:14:24 2009 From: Abdelrahman.Elhaddad at bibalex.org (Abdelrahman Elhaddad) Date: Sun, 25 Jan 2009 16:14:24 +0200 Subject: [slicer-devel] vtkSlicerViewerWidget::RequestRender() - Question? Message-ID: Dear Slicer Developers, I'm trying to create a "DVtkRenderWindow" which inherits from "vtkRenderWindow" and is defined in DIVERSE library, and use it as the main viewer instead of the "vtkRenderWindow" used in "vtkKWRenderWindget". I have a question regarding "RequestRender()" function in class "vtkSlcierViewerWidget". Basically, I need to know why do we need to request render instead of rendering directly. In other words why do we have to wait until the event loop is idle? Thank you. Abdelrahman Elhaddad Software Engineer Bibliotheca Alexandrina International School of Information Sciences (http://bibalex.org/isis/) VISTA (http://vista.bibalex.org) Tel.: +2034839999 - Ext.: 1436 Cell: +20106162660 From fedorov at bwh.harvard.edu Sun Jan 25 10:41:17 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sun, 25 Jan 2009 10:41:17 -0500 Subject: [slicer-devel] Help with testing Message-ID: <82f536810901250741o1c7ca886n5918acca296002a4@mail.gmail.com> Hi, I am trying to follow the testing framework in Applications/CLI to add testing to my loadable ChangeTracker module. For the beginning, I am trying to test just the logic, and for that I have a command line module corresponding to the logic in GUI module. The problem I have is that the test doesn't compile on Windows. I've been trying to resolve this for a couple of days now, but no luck. My test is in the trunk, Slicer3/Modules/ChangeTracker/Testing/ChangeTrackerTest.cxx, very simple: #if defined(_MSC_VER) #pragma warning ( disable : 4786 ) #endif #ifdef __BORLANDC__ #define ITK_LEAN_AND_MEAN #endif #include #include "itkTestMain.h" void RegisterTests() { REGISTER_TEST(ChangeTrackerTest); } #undef main #define main ChangeTrackerTest #include "ChangeTrackerCommandLine.cxx" It compiles and links on Linux, but on Windows I have this linker error: Linking... Creating library C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.lib and object C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.exp ChangeTrackerTest.obj : error LNK2019: unresolved external symbol "int __cdecl ChangeTrackerTest(int,char * * const)" (?ChangeTrackerTest@@YAHHQAPAD at Z) referenced in function "void __cdecl RegisterTests(void)" (?RegisterTests@@YAXXZ) C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.exe : fatal error LNK1120: 1 unresolved externals This doesn't make sense to me, because "main" is redefined to "ChangeTrackerTest", and main() *is* defined in ChangeTrackerCommandLine.cxx. Do you have any ideas why I have this error on Windows? I would greatly appreciate any help with this... Andriy Fedorov From bill.lorensen at gmail.com Sun Jan 25 13:00:23 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sun, 25 Jan 2009 13:00:23 -0500 Subject: [slicer-devel] Help with testing In-Reply-To: <82f536810901250741o1c7ca886n5918acca296002a4@mail.gmail.com> References: <82f536810901250741o1c7ca886n5918acca296002a4@mail.gmail.com> Message-ID: <4db4735c0901251000q273d4d57v2159b9a6c13491e2@mail.gmail.com> Andriy , The signature for your main program must be: int main(int argc, char * argv[]) not int main(int argc, char ** argv) Windows is pickier than linux. Bill On Sun, Jan 25, 2009 at 10:41 AM, Andriy Fedorov wrote: > Hi, > > I am trying to follow the testing framework in Applications/CLI to add > testing to my loadable ChangeTracker module. For the beginning, I am > trying to test just the logic, and for that I have a command line > module corresponding to the logic in GUI module. > > The problem I have is that the test doesn't compile on Windows. I've > been trying to resolve this for a couple of days now, but no luck. > > My test is in the trunk, > Slicer3/Modules/ChangeTracker/Testing/ChangeTrackerTest.cxx, very > simple: > > #if defined(_MSC_VER) > #pragma warning ( disable : 4786 ) > #endif > > #ifdef __BORLANDC__ > #define ITK_LEAN_AND_MEAN > #endif > > > #include > #include "itkTestMain.h" > void RegisterTests() > { > REGISTER_TEST(ChangeTrackerTest); > } > #undef main > #define main ChangeTrackerTest > #include "ChangeTrackerCommandLine.cxx" > > It compiles and links on Linux, but on Windows I have this linker error: > > Linking... > Creating library > C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.lib > and object C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.exp > ChangeTrackerTest.obj : error LNK2019: unresolved external symbol "int > __cdecl ChangeTrackerTest(int,char * * const)" > (?ChangeTrackerTest@@YAHHQAPAD at Z) referenced in function "void __cdecl > RegisterTests(void)" (?RegisterTests@@YAXXZ) > C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.exe : > fatal error LNK1120: 1 unresolved externals > > This doesn't make sense to me, because "main" is redefined to > "ChangeTrackerTest", and main() *is* defined in > ChangeTrackerCommandLine.cxx. > > Do you have any ideas why I have this error on Windows? > > I would greatly appreciate any help with this... > > Andriy Fedorov > From fedorov at bwh.harvard.edu Sun Jan 25 13:52:57 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sun, 25 Jan 2009 13:52:57 -0500 Subject: [slicer-devel] Help with testing In-Reply-To: <4db4735c0901251000q273d4d57v2159b9a6c13491e2@mail.gmail.com> References: <82f536810901250741o1c7ca886n5918acca296002a4@mail.gmail.com> <4db4735c0901251000q273d4d57v2159b9a6c13491e2@mail.gmail.com> Message-ID: <82f536810901251052l72d5dd6atf76ba3c7c259a649@mail.gmail.com> Thanks a lot, Bill. This was it. On Sun, Jan 25, 2009 at 1:00 PM, Bill Lorensen wrote: > Andriy , > > The signature for your main program must be: > int main(int argc, char * argv[]) > not > int main(int argc, char ** argv) > > Windows is pickier than linux. > > Bill > > > On Sun, Jan 25, 2009 at 10:41 AM, Andriy Fedorov > wrote: >> Hi, >> >> I am trying to follow the testing framework in Applications/CLI to add >> testing to my loadable ChangeTracker module. For the beginning, I am >> trying to test just the logic, and for that I have a command line >> module corresponding to the logic in GUI module. >> >> The problem I have is that the test doesn't compile on Windows. I've >> been trying to resolve this for a couple of days now, but no luck. >> >> My test is in the trunk, >> Slicer3/Modules/ChangeTracker/Testing/ChangeTrackerTest.cxx, very >> simple: >> >> #if defined(_MSC_VER) >> #pragma warning ( disable : 4786 ) >> #endif >> >> #ifdef __BORLANDC__ >> #define ITK_LEAN_AND_MEAN >> #endif >> >> >> #include >> #include "itkTestMain.h" >> void RegisterTests() >> { >> REGISTER_TEST(ChangeTrackerTest); >> } >> #undef main >> #define main ChangeTrackerTest >> #include "ChangeTrackerCommandLine.cxx" >> >> It compiles and links on Linux, but on Windows I have this linker error: >> >> Linking... >> Creating library >> C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.lib >> and object C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.exp >> ChangeTrackerTest.obj : error LNK2019: unresolved external symbol "int >> __cdecl ChangeTrackerTest(int,char * * const)" >> (?ChangeTrackerTest@@YAHHQAPAD at Z) referenced in function "void __cdecl >> RegisterTests(void)" (?RegisterTests@@YAXXZ) >> C:\cygwin\home\Andrey\Slicer3-build\bin\Debug\ChangeTrackerTest.exe : >> fatal error LNK1120: 1 unresolved externals >> >> This doesn't make sense to me, because "main" is redefined to >> "ChangeTrackerTest", and main() *is* defined in >> ChangeTrackerCommandLine.cxx. >> >> Do you have any ideas why I have this error on Windows? >> >> I would greatly appreciate any help with this... >> >> Andriy Fedorov >> > From pieper at bwh.harvard.edu Sun Jan 25 16:05:32 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sun, 25 Jan 2009 16:05:32 -0500 Subject: [slicer-devel] vtkSlicerViewerWidget::RequestRender() - Question? In-Reply-To: References: Message-ID: <497CD41C.8010001@bwh.harvard.edu> Hi Abdelrahman - The RequestRender method exists so that there is only a single actual render when multiple pieces of code modify the rendering pipeline (typically in response to an event). It basically allows the Render to take place synchronous with the window system events, rather than with taking place in-line as part of the callbacks. -Steve Abdelrahman Elhaddad wrote: > Dear Slicer Developers, > > I'm trying to create a "DVtkRenderWindow" which inherits from "vtkRenderWindow" and is defined in DIVERSE library, and use it as the main viewer instead of the "vtkRenderWindow" used in "vtkKWRenderWindget". I have a question regarding "RequestRender()" function in class "vtkSlcierViewerWidget". Basically, I need to know why do we need to request render instead of rendering directly. In other words why do we have to wait until the event loop is idle? Thank you. > > Abdelrahman Elhaddad > Software Engineer > Bibliotheca Alexandrina > International School of Information Sciences (http://bibalex.org/isis/) > VISTA (http://vista.bibalex.org) > Tel.: +2034839999 - Ext.: 1436 > Cell: +20106162660 > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From demian.wassermann at sophia.inria.fr Mon Jan 26 02:23:19 2009 From: demian.wassermann at sophia.inria.fr (Demian Wassermann) Date: Mon, 26 Jan 2009 08:23:19 +0100 Subject: [slicer-devel] A bit of help with tkinter and the python console Message-ID: <595E72D6-6E14-4015-9094-86DBC25475FD@sophia.inria.fr> Hi Luca, Steve, Daniel and all, Does anyone know why the stdout and stderr streams on the pyhton console do not have the flush operation? Can some TkInter/Tk wizard help me add it, I think that with this small patch I'll be able to enable a much much better console mode for python and exploratory analysis in Slicer Cheers Dem -- Demian Wassermann demian.wassermann at sophia.inria.fr PhD Student Odyssee Research Project INRIA Sophia-Antipolis 2004 route des lucioles - FR-06902 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090126/2bd2c1bf/attachment-0001.html From luca.antiga at gmail.com Mon Jan 26 05:47:53 2009 From: luca.antiga at gmail.com (Luca Antiga) Date: Mon, 26 Jan 2009 11:47:53 +0100 Subject: [slicer-devel] A bit of help with tkinter and the python console In-Reply-To: <595E72D6-6E14-4015-9094-86DBC25475FD@sophia.inria.fr> References: <595E72D6-6E14-4015-9094-86DBC25475FD@sophia.inria.fr> Message-ID: Hey Demian, that's because stdout is replaced by the OutputPipe class in Slicer3/ Base/GUI/Python/Console.py, which in turn uses this method def write(self, data, tag=None): """Show output from stdout or stderr in the console.""" if self.intraceback and data[-2:] == "\n ": data = data[:-1] start = self.text.index("insert") self.text.insert("insert", data) end = self.text.index("insert") if tag: self.text.tag_add(tag, start, end) from the Console class to write data to the 'fake' stdout. Since in this case there are no buffers involved (as far as I know), flushing doesn't make a lot of sense in my opinion. Were you having trouble with needing flushing or with the flush method not being there? If you need flush() to exist, we can add an empy flush method so that the interpreter will not complain when flush() is there. What do you think? Luca On Jan 26, 2009, at 8:23 AM, Demian Wassermann wrote: > Hi Luca, Steve, Daniel and all, > > Does anyone know why the stdout and stderr streams on the pyhton > console do not have the flush operation? > > Can some TkInter/Tk wizard help me add it, I think that with this > small patch I'll be able to enable a much much better console mode > for python and exploratory analysis in Slicer > > > Cheers > Dem > > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090126/7d00cec0/attachment.html From pieper at bwh.harvard.edu Mon Jan 26 08:27:48 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 26 Jan 2009 08:27:48 -0500 Subject: [slicer-devel] A bit of help with tkinter and the python console In-Reply-To: References: <595E72D6-6E14-4015-9094-86DBC25475FD@sophia.inria.fr> Message-ID: <497DBA54.6040501@bwh.harvard.edu> Hi Demian - Maybe the issue is that the tk side hasn't drawn yet - what if you do self.update() ? -Steve Luca Antiga wrote: > Hey Demian, > that's because stdout is replaced by the OutputPipe class in > Slicer3/Base/GUI/Python/Console.py, which in turn uses this method > def write(self, data, tag=None): > """Show output from stdout or stderr in the console.""" > if self.intraceback and data[-2:] == "\n ": data = data[:-1] > start = self.text.index("insert") > self.text.insert("insert", data) > end = self.text.index("insert") > if tag: self.text.tag_add(tag, start, end) > from the Console class to write data to the 'fake' stdout. Since in this > case there are no buffers involved (as far as I know), flushing doesn't > make a lot of sense in my opinion. > Were you having trouble with needing flushing or with the flush method > not being there? > If you need flush() to exist, we can add an empy flush method so that > the interpreter will not complain when flush() is there. > What do you think? > > > Luca > > > > > > On Jan 26, 2009, at 8:23 AM, Demian Wassermann wrote: > >> Hi Luca, Steve, Daniel and all, >> >> Does anyone know why the stdout and stderr streams on the pyhton >> console do not have the flush operation? >> >> Can some TkInter/Tk wizard help me add it, I think that with this >> small patch I'll be able to enable a much much better console mode for >> python and exploratory analysis in Slicer >> >> >> Cheers >> Dem >> >> >> -- >> Demian Wassermann >> demian.wassermann at sophia.inria.fr >> >> PhD Student >> Odyssee Research Project >> INRIA Sophia-Antipolis >> 2004 route des lucioles - FR-06902 >> From demian.wassermann at sophia.inria.fr Mon Jan 26 08:28:42 2009 From: demian.wassermann at sophia.inria.fr (Demian Wassermann) Date: Mon, 26 Jan 2009 14:28:42 +0100 Subject: [slicer-devel] A bit of help with tkinter and the python console In-Reply-To: <497DBA54.6040501@bwh.harvard.edu> References: <595E72D6-6E14-4015-9094-86DBC25475FD@sophia.inria.fr> <497DBA54.6040501@bwh.harvard.edu> Message-ID: mmmmmm not, is not that, I'm looking to it to check if I can use Ipython there D On Jan 26, 2009, at 2:27 PM, Steve Pieper wrote: > Hi Demian - > > Maybe the issue is that the tk side hasn't drawn yet - what if you > do self.update() ? > > -Steve > > Luca Antiga wrote: >> Hey Demian, >> that's because stdout is replaced by the OutputPipe class in >> Slicer3/Base/GUI/Python/Console.py, which in turn uses this method >> def write(self, data, tag=None): >> """Show output from stdout or stderr in the console.""" >> if self.intraceback and data[-2:] == "\n ": data = data[:-1] >> start = self.text.index("insert") >> self.text.insert("insert", data) >> end = self.text.index("insert") >> if tag: self.text.tag_add(tag, start, end) >> from the Console class to write data to the 'fake' stdout. Since >> in this case there are no buffers involved (as far as I know), >> flushing doesn't make a lot of sense in my opinion. >> Were you having trouble with needing flushing or with the flush >> method not being there? >> If you need flush() to exist, we can add an empy flush method so >> that the interpreter will not complain when flush() is there. >> What do you think? >> Luca >> On Jan 26, 2009, at 8:23 AM, Demian Wassermann wrote: >>> Hi Luca, Steve, Daniel and all, >>> >>> Does anyone know why the stdout and stderr streams on the pyhton >>> console do not have the flush operation? >>> Can some TkInter/Tk wizard help me add it, I think that with this >>> small patch I'll be able to enable a much much better console >>> mode for python and exploratory analysis in Slicer >>> >>> >>> Cheers >>> Dem >>> >>> >>> -- >>> Demian Wassermann >>> demian.wassermann at sophia.inria.fr >>> >>> PhD Student Odyssee Research Project >>> INRIA Sophia-Antipolis >>> 2004 route des lucioles - FR-06902 From pieper at bwh.harvard.edu Mon Jan 26 08:33:18 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 26 Jan 2009 08:33:18 -0500 Subject: [slicer-devel] A bit of help with tkinter and the python console In-Reply-To: References: <595E72D6-6E14-4015-9094-86DBC25475FD@sophia.inria.fr> <497DBA54.6040501@bwh.harvard.edu> Message-ID: <497DBB9E.6010705@bwh.harvard.edu> I got ipython to load once by hacking around on this but it didn't work correctly: http://ipython.scipy.org/moin/Cookbook/EmbeddingInTkinter Good luck, Steve Demian Wassermann wrote: > mmmmmm not, is not that, I'm looking to it to check if I can use Ipython > there > > > D > > > > On Jan 26, 2009, at 2:27 PM, Steve Pieper wrote: > >> Hi Demian - >> >> Maybe the issue is that the tk side hasn't drawn yet - what if you do >> self.update() ? >> >> -Steve >> >> Luca Antiga wrote: >>> Hey Demian, >>> that's because stdout is replaced by the OutputPipe class in >>> Slicer3/Base/GUI/Python/Console.py, which in turn uses this method >>> def write(self, data, tag=None): >>> """Show output from stdout or stderr in the console.""" >>> if self.intraceback and data[-2:] == "\n ": data = data[:-1] >>> start = self.text.index("insert") >>> self.text.insert("insert", data) >>> end = self.text.index("insert") >>> if tag: self.text.tag_add(tag, start, end) >>> from the Console class to write data to the 'fake' stdout. Since in >>> this case there are no buffers involved (as far as I know), flushing >>> doesn't make a lot of sense in my opinion. >>> Were you having trouble with needing flushing or with the flush >>> method not being there? >>> If you need flush() to exist, we can add an empy flush method so that >>> the interpreter will not complain when flush() is there. >>> What do you think? >>> Luca >>> On Jan 26, 2009, at 8:23 AM, Demian Wassermann wrote: >>>> Hi Luca, Steve, Daniel and all, >>>> >>>> Does anyone know why the stdout and stderr streams on the pyhton >>>> console do not have the flush operation? >>>> Can some TkInter/Tk wizard help me add it, I think that with this >>>> small patch I'll be able to enable a much much better console mode >>>> for python and exploratory analysis in Slicer >>>> >>>> >>>> Cheers >>>> Dem >>>> >>>> >>>> -- >>>> Demian Wassermann >>>> demian.wassermann at sophia.inria.fr >>>> >>>> PhD Student Odyssee Research Project >>>> INRIA Sophia-Antipolis >>>> 2004 route des lucioles - FR-06902 > From luca.antiga at gmail.com Mon Jan 26 08:41:25 2009 From: luca.antiga at gmail.com (Luca Antiga) Date: Mon, 26 Jan 2009 14:41:25 +0100 Subject: [slicer-devel] A bit of help with tkinter and the python console In-Reply-To: References: <595E72D6-6E14-4015-9094-86DBC25475FD@sophia.inria.fr> <497DBA54.6040501@bwh.harvard.edu> Message-ID: Hey, I desume I wasn't getting the point. I think I see it now. Indeed it would be great to have a iPython-enabled console. Steve, the code you posted looks like a full replacement for Console.py. No question we should give it a try. Luca On Jan 26, 2009, at 2:28 PM, Demian Wassermann wrote: > mmmmmm not, is not that, I'm looking to it to check if I can use > Ipython there > > > D > > > > On Jan 26, 2009, at 2:27 PM, Steve Pieper wrote: > >> Hi Demian - >> >> Maybe the issue is that the tk side hasn't drawn yet - what if you >> do self.update() ? >> >> -Steve >> >> Luca Antiga wrote: >>> Hey Demian, >>> that's because stdout is replaced by the OutputPipe class in >>> Slicer3/Base/GUI/Python/Console.py, which in turn uses this method >>> def write(self, data, tag=None): >>> """Show output from stdout or stderr in the console.""" >>> if self.intraceback and data[-2:] == "\n ": data = data[:-1] >>> start = self.text.index("insert") >>> self.text.insert("insert", data) >>> end = self.text.index("insert") >>> if tag: self.text.tag_add(tag, start, end) >>> from the Console class to write data to the 'fake' stdout. Since >>> in this case there are no buffers involved (as far as I know), >>> flushing doesn't make a lot of sense in my opinion. >>> Were you having trouble with needing flushing or with the flush >>> method not being there? >>> If you need flush() to exist, we can add an empy flush method so >>> that the interpreter will not complain when flush() is there. >>> What do you think? >>> Luca >>> On Jan 26, 2009, at 8:23 AM, Demian Wassermann wrote: >>>> Hi Luca, Steve, Daniel and all, >>>> >>>> Does anyone know why the stdout and stderr streams on the pyhton >>>> console do not have the flush operation? >>>> Can some TkInter/Tk wizard help me add it, I think that with this >>>> small patch I'll be able to enable a much much better console >>>> mode for python and exploratory analysis in Slicer >>>> >>>> >>>> Cheers >>>> Dem >>>> >>>> >>>> -- >>>> Demian Wassermann >>>> demian.wassermann at sophia.inria.fr >>> > >>>> PhD Student Odyssee Research Project >>>> INRIA Sophia-Antipolis >>>> 2004 route des lucioles - FR-06902 > -- Luca Antiga, PhD Head, Medical Imaging Unit, Biomedical Engineering Department, Mario Negri Institute. mail: Villa Camozzi, 24020, Ranica (BG), Italy phone: +39 035 4535-381 email: antiga at marionegri.it web: http://villacamozzi.marionegri.it/~luca -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090126/67013b92/attachment-0001.html From simon at math.uh.edu Mon Jan 26 11:22:37 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Mon, 26 Jan 2009 10:22:37 -0600 Subject: [slicer-devel] build failure on OS X Message-ID: <200901261022.38153.simon@math.uh.edu> Hi, Has something recently broken the Mac OS X build? My build fails on BMString.cxx in batch make utilities.. cheers, Simon From pieper at bwh.harvard.edu Mon Jan 26 12:36:17 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 26 Jan 2009 12:36:17 -0500 Subject: [slicer-devel] build failure on OS X In-Reply-To: <200901261022.38153.simon@math.uh.edu> References: <200901261022.38153.simon@math.uh.edu> Message-ID: <497DF491.2050405@bwh.harvard.edu> Hi Simon - It looks like BMString is a new class that is breaking all the BatchMake builds: http://www.cdash.org/CDash/index.php?project=BatchMake As a workaround for your slicer build, you can back up to an older version of BatchMake with something like this: % cd Slicer3-lib/BatchMake % cvs -q update -D 2009-01-22 % cd ../BatchMake-build/ % make Best, Steve Simon K Alexander wrote: > Hi, > > Has something recently broken the Mac OS X build? My build fails on BMString.cxx in batch make utilities.. > > cheers, > Simon > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From julien.finet at kitware.com Mon Jan 26 12:41:52 2009 From: julien.finet at kitware.com (Julien Finet) Date: Mon, 26 Jan 2009 12:41:52 -0500 Subject: [slicer-devel] build failure on OS X In-Reply-To: <497DF491.2050405@bwh.harvard.edu> References: <200901261022.38153.simon@math.uh.edu> <497DF491.2050405@bwh.harvard.edu> Message-ID: <5b9fa8230901260941q58029f99wfae669434ab640e0@mail.gmail.com> I've just fixed the compilation error, sorry about that. Let me know if you still have some issues with it. Regards, Julien. On Mon, Jan 26, 2009 at 12:36 PM, Steve Pieper wrote: > Hi Simon - > > It looks like BMString is a new class that is breaking all the BatchMake > builds: > > http://www.cdash.org/CDash/index.php?project=BatchMake > > > As a workaround for your slicer build, you can back up to an older > version of BatchMake with something like this: > > % cd Slicer3-lib/BatchMake > % cvs -q update -D 2009-01-22 > % cd ../BatchMake-build/ > % make > > Best, > Steve > > > > Simon K Alexander wrote: > > Hi, > > > > Has something recently broken the Mac OS X build? My build fails on > BMString.cxx in batch make utilities.. > > > > cheers, > > Simon > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090126/cb21ce48/attachment.html From yumin.yuan at kitware.com Mon Jan 26 15:25:32 2009 From: yumin.yuan at kitware.com (Yumin Yuan) Date: Mon, 26 Jan 2009 15:25:32 -0500 Subject: [slicer-devel] build failure on OS X In-Reply-To: <5b9fa8230901260941q58029f99wfae669434ab640e0@mail.gmail.com> References: <200901261022.38153.simon@math.uh.edu> <497DF491.2050405@bwh.harvard.edu> <5b9fa8230901260941q58029f99wfae669434ab640e0@mail.gmail.com> Message-ID: Hi Julien, I still have some build errors on my linux. /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx: In member function 'BMString& BMString::operator=(const char*)': /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx:246: error: 'strlen' was not declared in this scope /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx: In member function 'BMString& BMString::replace(const char*, const char*)': /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx:511: error: 'strlen' was not declared in this scope /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx: In member function 'BMString BMString::replaceCopy(const char*, const char*) const': /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx:526: error: 'strlen' was not declared in this scope Yumin On Mon, Jan 26, 2009 at 12:41 PM, Julien Finet wrote: > I've just fixed the compilation error, sorry about that. Let me know if you > still have some issues with it. > Regards, > Julien. > > On Mon, Jan 26, 2009 at 12:36 PM, Steve Pieper > wrote: >> >> Hi Simon - >> >> It looks like BMString is a new class that is breaking all the BatchMake >> builds: >> >> http://www.cdash.org/CDash/index.php?project=BatchMake >> >> >> As a workaround for your slicer build, you can back up to an older >> version of BatchMake with something like this: >> >> % cd Slicer3-lib/BatchMake >> % cvs -q update -D 2009-01-22 >> % cd ../BatchMake-build/ >> % make >> >> Best, >> Steve >> >> >> >> Simon K Alexander wrote: >> > Hi, >> > >> > Has something recently broken the Mac OS X build? My build fails on >> > BMString.cxx in batch make utilities.. >> > >> > cheers, >> > Simon >> > _______________________________________________ >> > slicer-devel mailing list >> > slicer-devel at bwh.harvard.edu >> > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> > To unsubscribe: send email to >> > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> > subject >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > -- ~~~~~~~~~~~~~~~~ Yumin Yuan R&D Engineer Kitware Inc. www.kitware.com ~~~~~~~~~~~~~~~~ From julien.finet at kitware.com Mon Jan 26 15:35:07 2009 From: julien.finet at kitware.com (Julien Finet) Date: Mon, 26 Jan 2009 15:35:07 -0500 Subject: [slicer-devel] build failure on OS X In-Reply-To: References: <200901261022.38153.simon@math.uh.edu> <497DF491.2050405@bwh.harvard.edu> <5b9fa8230901260941q58029f99wfae669434ab640e0@mail.gmail.com> Message-ID: <5b9fa8230901261235t4a01f2fbo97b485b352a2bb55@mail.gmail.com> Hi Yumin, It should be fixed now. Thanks for your patience, Julien. On Mon, Jan 26, 2009 at 3:25 PM, Yumin Yuan wrote: > Hi Julien, > I still have some build errors on my linux. > > /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx: In > member function 'BMString& BMString::operator=(const char*)': > /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx:246: > error: 'strlen' was not declared in this scope > /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx: In > member function 'BMString& BMString::replace(const char*, const > char*)': > /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx:511: > error: 'strlen' was not declared in this scope > /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx: In > member function 'BMString BMString::replaceCopy(const char*, const > char*) const': > /home/yumin/Slicer/Slicer3-lib/BatchMake/Utilities/BMString.cxx:526: > error: 'strlen' was not declared in this scope > > Yumin > > On Mon, Jan 26, 2009 at 12:41 PM, Julien Finet > wrote: > > I've just fixed the compilation error, sorry about that. Let me know if > you > > still have some issues with it. > > Regards, > > Julien. > > > > On Mon, Jan 26, 2009 at 12:36 PM, Steve Pieper > > wrote: > >> > >> Hi Simon - > >> > >> It looks like BMString is a new class that is breaking all the BatchMake > >> builds: > >> > >> http://www.cdash.org/CDash/index.php?project=BatchMake > >> > >> > >> As a workaround for your slicer build, you can back up to an older > >> version of BatchMake with something like this: > >> > >> % cd Slicer3-lib/BatchMake > >> % cvs -q update -D 2009-01-22 > >> % cd ../BatchMake-build/ > >> % make > >> > >> Best, > >> Steve > >> > >> > >> > >> Simon K Alexander wrote: > >> > Hi, > >> > > >> > Has something recently broken the Mac OS X build? My build fails on > >> > BMString.cxx in batch make utilities.. > >> > > >> > cheers, > >> > Simon > >> > _______________________________________________ > >> > slicer-devel mailing list > >> > slicer-devel at bwh.harvard.edu > >> > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > >> > To unsubscribe: send email to > >> > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > >> > subject > >> _______________________________________________ > >> slicer-devel mailing list > >> slicer-devel at bwh.harvard.edu > >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > >> To unsubscribe: send email to > >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > >> subject > > > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu > > with unsubscribe as the subject > > > > > > -- > ~~~~~~~~~~~~~~~~ > Yumin Yuan > R&D Engineer > Kitware Inc. > www.kitware.com > ~~~~~~~~~~~~~~~~ > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090126/278b5f0b/attachment.html From fedorov at bwh.harvard.edu Tue Jan 27 11:52:03 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 27 Jan 2009 11:52:03 -0500 Subject: [slicer-devel] Problems with Windows installation Message-ID: <82f536810901270852p1f0d2752y480ccc226342e8e0@mail.gmail.com> Hi, I have a user (Jean-Jacques) who has problems with nightly Slicer build. He has an older version, which he uninstalled, but the new one doesn't run with the message "::Slicer3_REAL": no such variable" (also see the screenshot attached). This is Windows Vista. Could it be that there are some leftovers of the previous installation in the registry, that mess up the environment? We made sure the previous installation was uninstalled and physically removed from disk, and the machine was rebooted before the new installation. Any ideas how this can be fixed? Thanks Andriy Fedorov -------------- next part -------------- A non-text attachment was scrubbed... Name: bug.png Type: image/png Size: 43951 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090127/0b581363/attachment-0001.png From sebastien.barre at kitware.com Tue Jan 27 11:55:58 2009 From: sebastien.barre at kitware.com (Sebastien BARRE) Date: Tue, 27 Jan 2009 11:55:58 -0500 Subject: [slicer-devel] Problems with Windows installation In-Reply-To: <82f536810901270852p1f0d2752y480ccc226342e8e0@mail.gmail.co m> References: <82f536810901270852p1f0d2752y480ccc226342e8e0@mail.gmail.com> Message-ID: <20090127165642.DFE5548DC@public.kitware.com> At 1/27/2009 11:52 AM, Andriy Fedorov wrote: >Hi, > >I have a user (Jean-Jacques) who has problems with nightly Slicer >build. He has an older version, which he uninstalled, but the new one >doesn't run with the message "::Slicer3_REAL": no such variable" (also >see the screenshot attached). This is Windows Vista. > >Could it be that there are some leftovers of the previous installation >in the registry, that mess up the environment? No, the launch.tcl launcher has been broken for several months. Try executing the other launcher inside the bin/ directory, also called Slicer3.exe From nagy.attila at yahoo.com Tue Jan 27 11:58:27 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Tue, 27 Jan 2009 08:58:27 -0800 (PST) Subject: [slicer-devel] Problems with Windows installation In-Reply-To: <82f536810901270852p1f0d2752y480ccc226342e8e0@mail.gmail.com> Message-ID: <744698.35627.qm@web37402.mail.mud.yahoo.com> Hope it helps: I just installed the very same build (nightly 22-01) on an XP Professional, yesterday, and it had no such problems. This was the very first Slicer install on that machine, so we had no leftovers for sure. Attila --- On Tue, 1/27/09, Andriy Fedorov wrote: > From: Andriy Fedorov > Subject: [slicer-devel] Problems with Windows installation > To: slicer-devel at bwh.harvard.edu, slicer-users at bwh.harvard.edu > Cc: "Jean-Jacques Lemaire" > Date: Tuesday, January 27, 2009, 5:52 PM > Hi, > > I have a user (Jean-Jacques) who has problems with nightly > Slicer > build. He has an older version, which he uninstalled, but > the new one > doesn't run with the message > "::Slicer3_REAL": no such variable" (also > see the screenshot attached). This is Windows Vista. > > Could it be that there are some leftovers of the previous > installation > in the registry, that mess up the environment? We made sure > the > previous installation was uninstalled and physically > removed from > disk, and the machine was rebooted before the new > installation. > > Any ideas how this can be fixed? > > Thanks > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with > unsubscribe as the subject From fedorov at bwh.harvard.edu Tue Jan 27 12:00:53 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 27 Jan 2009 12:00:53 -0500 Subject: [slicer-devel] Problems with Windows installation In-Reply-To: <744698.35627.qm@web37402.mail.mud.yahoo.com> References: <82f536810901270852p1f0d2752y480ccc226342e8e0@mail.gmail.com> <744698.35627.qm@web37402.mail.mud.yahoo.com> Message-ID: <82f536810901270900o109cf31ejecee5ab9dba476fa@mail.gmail.com> We did try Slicer3.exe from bin, it doesn't work -- there problems finding dll's. The same installation works just fine on my Win XP. On Tue, Jan 27, 2009 at 11:58 AM, Attila Nagy wrote: > Hope it helps: > I just installed the very same build (nightly 22-01) on an XP Professional, yesterday, and it had no such problems. > This was the very first Slicer install on that machine, so we had no leftovers for sure. > > Attila > > --- On Tue, 1/27/09, Andriy Fedorov wrote: > >> From: Andriy Fedorov >> Subject: [slicer-devel] Problems with Windows installation >> To: slicer-devel at bwh.harvard.edu, slicer-users at bwh.harvard.edu >> Cc: "Jean-Jacques Lemaire" >> Date: Tuesday, January 27, 2009, 5:52 PM >> Hi, >> >> I have a user (Jean-Jacques) who has problems with nightly >> Slicer >> build. He has an older version, which he uninstalled, but >> the new one >> doesn't run with the message >> "::Slicer3_REAL": no such variable" (also >> see the screenshot attached). This is Windows Vista. >> >> Could it be that there are some leftovers of the previous >> installation >> in the registry, that mess up the environment? We made sure >> the >> previous installation was uninstalled and physically >> removed from >> disk, and the machine was rebooted before the new >> installation. >> >> Any ideas how this can be fixed? >> >> Thanks >> >> Andriy Fedorov >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with >> unsubscribe as the subject > > > > From pieper at bwh.harvard.edu Tue Jan 27 12:27:03 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 27 Jan 2009 12:27:03 -0500 Subject: [slicer-devel] [slicer-users] Problems with Windows installation In-Reply-To: <82f536810901270900o109cf31ejecee5ab9dba476fa@mail.gmail.com> References: <82f536810901270852p1f0d2752y480ccc226342e8e0@mail.gmail.com> <744698.35627.qm@web37402.mail.mud.yahoo.com> <82f536810901270900o109cf31ejecee5ab9dba476fa@mail.gmail.com> Message-ID: <497F43E7.10406@bwh.harvard.edu> Hi Andriy - Is this machine someplace where I could have a look at it? -Steve Andriy Fedorov wrote: > We did try Slicer3.exe from bin, it doesn't work -- there problems > finding dll's. > > The same installation works just fine on my Win XP. > > > On Tue, Jan 27, 2009 at 11:58 AM, Attila Nagy wrote: >> Hope it helps: >> I just installed the very same build (nightly 22-01) on an XP Professional, yesterday, and it had no such problems. >> This was the very first Slicer install on that machine, so we had no leftovers for sure. >> >> Attila >> >> --- On Tue, 1/27/09, Andriy Fedorov wrote: >> >>> From: Andriy Fedorov >>> Subject: [slicer-devel] Problems with Windows installation >>> To: slicer-devel at bwh.harvard.edu, slicer-users at bwh.harvard.edu >>> Cc: "Jean-Jacques Lemaire" >>> Date: Tuesday, January 27, 2009, 5:52 PM >>> Hi, >>> >>> I have a user (Jean-Jacques) who has problems with nightly >>> Slicer >>> build. He has an older version, which he uninstalled, but >>> the new one >>> doesn't run with the message >>> "::Slicer3_REAL": no such variable" (also >>> see the screenshot attached). This is Windows Vista. >>> >>> Could it be that there are some leftovers of the previous >>> installation >>> in the registry, that mess up the environment? We made sure >>> the >>> previous installation was uninstalled and physically >>> removed from >>> disk, and the machine was rebooted before the new >>> installation. >>> >>> Any ideas how this can be fixed? >>> >>> Thanks >>> >>> Andriy Fedorov >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with >>> unsubscribe as the subject >> >> >> > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From nagy.attila at yahoo.com Tue Jan 27 13:32:57 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Tue, 27 Jan 2009 10:32:57 -0800 (PST) Subject: [slicer-devel] Batchmake packaging error Message-ID: <143061.71884.qm@web37408.mail.mud.yahoo.com> Hi all, I just caught this, when generating the package. I have deleted the Batchmake and Batchmake-build dirs before compilation. Attila CPack: - Install project: teem CMake Error at ../Slicer3-lib/BatchMake-build/Utilities/bmModuleDescriptionParser/cmake_install.cmake:66 (FILE): file INSTALL cannot find file "/ufsbckp/64-Slicer3.3/Slicer3-lib/BatchMake-build/Utilities/bmModuleDescriptionParser/UsebmModuleDescriptionParser.cmake" to install. Call Stack (most recent call first): ../Slicer3-lib/BatchMake-build/Utilities/cmake_install.cmake:65 (INCLUDE) ../Slicer3-lib/BatchMake-build/cmake_install.cmake:37 (INCLUDE) CPack Error: Error when generating package: Slicer3 gmake: *** [package] Error 1 From hayes at bwh.harvard.edu Tue Jan 27 14:23:45 2009 From: hayes at bwh.harvard.edu (Kathryn Hayes) Date: Tue, 27 Jan 2009 14:23:45 -0500 (EST) Subject: [slicer-devel] Batchmake packaging error In-Reply-To: <143061.71884.qm@web37408.mail.mud.yahoo.com> References: <143061.71884.qm@web37408.mail.mud.yahoo.com> Message-ID: I'm getting the following error on Linux 32 when trying to build BatchMake, which looks related: [ 9%] Building CXX object Utilities/bmModuleDescriptionParser/CMakeFiles/bmModuleDescriptionParser.dir/bmModuleProcessInformation.o /home/hayes/Slicer3-release-gbt/Slicer3-lib/BatchMake/Utilities/bmModuleDescriptionParser/bmModuleProcessInformation.cxx:14:38: ModuleProcessInformation.h: No such file or directory /home/hayes/Slicer3-release-gbt/Slicer3-lib/BatchMake/Utilities/bmModuleDescriptionParser/bmModuleProcessInformation.cxx:17: error: expected constructor, destructor, or type conversion before '&' token /home/hayes/Slicer3-release-gbt/Slicer3-lib/BatchMake/Utilities/bmModuleDescriptionParser/bmModuleProcessInformation.cxx:17: error: expected `,' or `;' before '&' token make[2]: *** [Utilities/bmModuleDescriptionParser/CMakeFiles/bmModuleDescriptionParser.dir/bmModuleProcessInformation.o] Error 1 make[1]: *** [Utilities/bmModuleDescriptionParser/CMakeFiles/bmModuleDescriptionParser.dir/all] Error 2 make[1]: *** Waiting for unfinished jobs.... [ 10%] Building CXX object Utilities/CMakeFiles/BatchMakeUtilities.dir/HttpRequest.o Linking CXX shared library ../bin/libBatchMakeUtilities.so make[1]: *** Waiting for unfinished jobs.... [ 10%] make[1]: *** Waiting for unfinished jobs.... Built target BatchMakeUtilities make: *** [all] Error 2 On Tue, 27 Jan 2009, Attila Nagy wrote: > Hi all, > > I just caught this, when generating the package. > I have deleted the Batchmake and Batchmake-build dirs before compilation. > > Attila > > > CPack: - Install project: teem > CMake Error at ../Slicer3-lib/BatchMake-build/Utilities/bmModuleDescriptionParser/cmake_install.cmake:66 (FILE): > file INSTALL cannot find file > "/ufsbckp/64-Slicer3.3/Slicer3-lib/BatchMake-build/Utilities/bmModuleDescriptionParser/UsebmModuleDescriptionParser.cmake" > to install. > Call Stack (most recent call first): > ../Slicer3-lib/BatchMake-build/Utilities/cmake_install.cmake:65 (INCLUDE) > ../Slicer3-lib/BatchMake-build/cmake_install.cmake:37 (INCLUDE) > > > CPack Error: Error when generating package: Slicer3 > gmake: *** [package] Error 1 > > > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From julien.finet at kitware.com Tue Jan 27 14:52:05 2009 From: julien.finet at kitware.com (Julien Finet) Date: Tue, 27 Jan 2009 14:52:05 -0500 Subject: [slicer-devel] Batchmake packaging error In-Reply-To: References: <143061.71884.qm@web37408.mail.mud.yahoo.com> Message-ID: <5b9fa8230901271152y4f8b045kf494dd8766ad0974@mail.gmail.com> Try to update now, Julien (Jomier) just fixed it. Regards, Julien. On Tue, Jan 27, 2009 at 2:23 PM, Kathryn Hayes wrote: > I'm getting the following error on Linux 32 when trying to build > BatchMake, which looks related: > > [ 9%] Building CXX object > > Utilities/bmModuleDescriptionParser/CMakeFiles/bmModuleDescriptionParser.dir/bmModuleProcessInformation.o > > /home/hayes/Slicer3-release-gbt/Slicer3-lib/BatchMake/Utilities/bmModuleDescriptionParser/bmModuleProcessInformation.cxx:14:38: > ModuleProcessInformation.h: No such file or directory > > /home/hayes/Slicer3-release-gbt/Slicer3-lib/BatchMake/Utilities/bmModuleDescriptionParser/bmModuleProcessInformation.cxx:17: > error: expected constructor, destructor, or type conversion before '&' > token > > /home/hayes/Slicer3-release-gbt/Slicer3-lib/BatchMake/Utilities/bmModuleDescriptionParser/bmModuleProcessInformation.cxx:17: > error: expected `,' or `;' before '&' token > make[2]: *** > > [Utilities/bmModuleDescriptionParser/CMakeFiles/bmModuleDescriptionParser.dir/bmModuleProcessInformation.o] > Error 1 > make[1]: *** > > [Utilities/bmModuleDescriptionParser/CMakeFiles/bmModuleDescriptionParser.dir/all] > Error 2 > make[1]: *** Waiting for unfinished jobs.... > [ 10%] Building CXX object > Utilities/CMakeFiles/BatchMakeUtilities.dir/HttpRequest.o > Linking CXX shared library ../bin/libBatchMakeUtilities.so > make[1]: *** Waiting for unfinished jobs.... > [ 10%] make[1]: *** Waiting for unfinished jobs.... > Built target BatchMakeUtilities > make: *** [all] Error 2 > > > On Tue, 27 Jan 2009, Attila Nagy wrote: > > > Hi all, > > > > I just caught this, when generating the package. > > I have deleted the Batchmake and Batchmake-build dirs before compilation. > > > > Attila > > > > > > CPack: - Install project: teem > > CMake Error at > ../Slicer3-lib/BatchMake-build/Utilities/bmModuleDescriptionParser/cmake_install.cmake:66 > (FILE): > > file INSTALL cannot find file > > > "/ufsbckp/64-Slicer3.3/Slicer3-lib/BatchMake-build/Utilities/bmModuleDescriptionParser/UsebmModuleDescriptionParser.cmake" > > to install. > > Call Stack (most recent call first): > > ../Slicer3-lib/BatchMake-build/Utilities/cmake_install.cmake:65 > (INCLUDE) > > ../Slicer3-lib/BatchMake-build/cmake_install.cmake:37 (INCLUDE) > > > > > > CPack Error: Error when generating package: Slicer3 > > gmake: *** [package] Error 1 > > > > > > > > > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090127/ba580143/attachment-0001.html From fedorov at bwh.harvard.edu Tue Jan 27 18:16:42 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 27 Jan 2009 18:16:42 -0500 Subject: [slicer-devel] Problems with Windows installation -- RESOLVED Message-ID: <82f536810901271516w147f657bs91c0e2fb86dbd55@mail.gmail.com> The problem appeared to be that Slicer installation folder (located in "Program Files") had "read-only" flag set. Vista worked hard, but couldn't revert this setting, ignoring our repeated attempts. After copying Slicer installation folder to a different location outside "Program Files", Slicer started correctly. Hope this helps somebody down the Slicer Vista road... On Tue, Jan 27, 2009 at 12:00 PM, Andriy Fedorov wrote: > We did try Slicer3.exe from bin, it doesn't work -- there problems > finding dll's. > > The same installation works just fine on my Win XP. > > > On Tue, Jan 27, 2009 at 11:58 AM, Attila Nagy wrote: >> Hope it helps: >> I just installed the very same build (nightly 22-01) on an XP Professional, yesterday, and it had no such problems. >> This was the very first Slicer install on that machine, so we had no leftovers for sure. >> >> Attila >> >> --- On Tue, 1/27/09, Andriy Fedorov wrote: >> >>> From: Andriy Fedorov >>> Subject: [slicer-devel] Problems with Windows installation >>> To: slicer-devel at bwh.harvard.edu, slicer-users at bwh.harvard.edu >>> Cc: "Jean-Jacques Lemaire" >>> Date: Tuesday, January 27, 2009, 5:52 PM >>> Hi, >>> >>> I have a user (Jean-Jacques) who has problems with nightly >>> Slicer >>> build. He has an older version, which he uninstalled, but >>> the new one >>> doesn't run with the message >>> "::Slicer3_REAL": no such variable" (also >>> see the screenshot attached). This is Windows Vista. >>> >>> Could it be that there are some leftovers of the previous >>> installation >>> in the registry, that mess up the environment? We made sure >>> the >>> previous installation was uninstalled and physically >>> removed from >>> disk, and the machine was rebooted before the new >>> installation. >>> >>> Any ideas how this can be fixed? >>> >>> Thanks >>> >>> Andriy Fedorov >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with >>> unsubscribe as the subject >> >> >> >> > From demian.wassermann at sophia.inria.fr Wed Jan 28 07:48:23 2009 From: demian.wassermann at sophia.inria.fr (Demian Wassermann) Date: Wed, 28 Jan 2009 13:48:23 +0100 Subject: [slicer-devel] OpenIGTLink Message-ID: <56B3E899-6283-4D88-9705-0FF610227C6E@sophia.inria.fr> Hi, how are you doing? I've found out that the OpenIGTLink module is stalling slicer!!!! I even deleted the module, set up the slicer_variables.tcl to "OFF" and cleaned the module cache And it still stalls and does a SIGSEGV Any help? Dem -- Demian Wassermann demian.wassermann at sophia.inria.fr PhD Student Odyssee Research Project INRIA Sophia-Antipolis 2004 route des lucioles - FR-06902 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090128/886b5264/attachment.html From tokuda at bwh.harvard.edu Wed Jan 28 08:31:42 2009 From: tokuda at bwh.harvard.edu (Junichi Tokuda) Date: Wed, 28 Jan 2009 08:31:42 -0500 Subject: [slicer-devel] OpenIGTLink In-Reply-To: <56B3E899-6283-4D88-9705-0FF610227C6E@sophia.inria.fr> References: <56B3E899-6283-4D88-9705-0FF610227C6E@sophia.inria.fr> Message-ID: Hi Demian, Could you tell me about more detail? What I would like to know are: - Your OS, Slicer revision, etc. - In which stage does the Slicer stall? - Which file did you delete? To delete the module, you need to delete Slicer3-build/lib/Slicer3/Modules/libOpenIGTLinkIF.{dll, so, dylib} Thanks, Junichi --- Junichi Tokuda, Ph.D. Department of Radiology Brigham and Women's Hospital Harvard Medical School 75 Francis Street, ASB-I L1-050 Boston, MA 02115, USA tokuda at bwh.harvard.edu On Jan 28, 2009, at 7:48 AM, Demian Wassermann wrote: > Hi, how are you doing? > > I've found out that the OpenIGTLink module is stalling slicer!!!! > > I even deleted the module, set up the slicer_variables.tcl to "OFF" > and cleaned the module cache > > And it still stalls and does a SIGSEGV > > Any help? > > Dem > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From hata at bwh.harvard.edu Wed Jan 28 08:49:36 2009 From: hata at bwh.harvard.edu (Nobuhiko Hata) Date: Wed, 28 Jan 2009 08:49:36 -0500 Subject: [slicer-devel] OpenIGTLink In-Reply-To: <56B3E899-6283-4D88-9705-0FF610227C6E@sophia.inria.fr> References: <56B3E899-6283-4D88-9705-0FF610227C6E@sophia.inria.fr> Message-ID: <49806270.3040508@bwh.harvard.edu> Hi Dem Which version of Slicer do you use with OpenIGTLink? Noby > Hi, how are you doing? > > I've found out that the OpenIGTLink module is stalling slicer!!!! > > I even deleted the module, set up the slicer_variables.tcl to "OFF" and > cleaned the module cache > > And it still stalls and does a SIGSEGV > > Any help? > > Dem > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From demian.wassermann at sophia.inria.fr Wed Jan 28 10:27:13 2009 From: demian.wassermann at sophia.inria.fr (Demian Wassermann) Date: Wed, 28 Jan 2009 16:27:13 +0100 Subject: [slicer-devel] OpenIGTLink In-Reply-To: <49806270.3040508@bwh.harvard.edu> References: <56B3E899-6283-4D88-9705-0FF610227C6E@sophia.inria.fr> <49806270.3040508@bwh.harvard.edu> Message-ID: <9ED3DC8C-B602-46DD-B3E8-AD8A0B3BBA1F@sophia.inria.fr> Hi Noby and Junichi I'm using Mac OSX 10.4 and I'm using the slicer head trunk version and I've built everything from scratch with getbuildtest.tcl Cheers D -- Demian Wassermann demian.wassermann at sophia.inria.fr PhD Student Odyssee Research Project INRIA Sophia-Antipolis 2004 route des lucioles - FR-06902 On Jan 28, 2009, at 2:49 PM, Nobuhiko Hata wrote: > Hi Dem > > Which version of Slicer do you use with OpenIGTLink? > > > Noby > >> Hi, how are you doing? >> I've found out that the OpenIGTLink module is stalling slicer!!!! >> I even deleted the module, set up the slicer_variables.tcl to >> "OFF" and cleaned the module cache >> And it still stalls and does a SIGSEGV >> Any help? >> Dem >> -- >> Demian Wassermann >> demian.wassermann at sophia.inria.fr >> >> PhD Student Odyssee Research Project >> INRIA Sophia-Antipolis >> 2004 route des lucioles - FR-06902 >> --------------------------------------------------------------------- >> --- >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel- >> request at massmail.spl.harvard.edu with unsubscribe as the subject From tokuda at bwh.harvard.edu Wed Jan 28 20:36:43 2009 From: tokuda at bwh.harvard.edu (Junichi Tokuda) Date: Wed, 28 Jan 2009 20:36:43 -0500 Subject: [slicer-devel] OpenIGTLink In-Reply-To: <9ED3DC8C-B602-46DD-B3E8-AD8A0B3BBA1F@sophia.inria.fr> References: <56B3E899-6283-4D88-9705-0FF610227C6E@sophia.inria.fr> <49806270.3040508@bwh.harvard.edu> <9ED3DC8C-B602-46DD-B3E8-AD8A0B3BBA1F@sophia.inria.fr> Message-ID: <973AC55C-A6EA-46BE-8C7B-42F2C6D1E248@bwh.harvard.edu> Hi Demian, I couldn't reproduce the error you got... Did you delete libOpenIGTLinkIF.dylib (or entire Slicer3-build directory) before running getbuildtest.tcl with USE_OPENIGTLINK="OFF"? If not, the old dylib file remains in the module directory and might cause segmentation fault. Junichi On Jan 28, 2009, at 10:27 AM, Demian Wassermann wrote: > Hi Noby and Junichi > > I'm using Mac OSX 10.4 and I'm using the slicer head trunk version > and I've built everything from scratch with getbuildtest.tcl > > Cheers > D > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > > > > On Jan 28, 2009, at 2:49 PM, Nobuhiko Hata wrote: > >> Hi Dem >> >> Which version of Slicer do you use with OpenIGTLink? >> >> >> Noby >> >>> Hi, how are you doing? >>> I've found out that the OpenIGTLink module is stalling slicer!!!! >>> I even deleted the module, set up the slicer_variables.tcl to >>> "OFF" and cleaned the module cache >>> And it still stalls and does a SIGSEGV >>> Any help? >>> Dem >>> -- >>> Demian Wassermann >>> demian.wassermann at sophia.inria.fr >> > >>> PhD Student Odyssee Research Project >>> INRIA Sophia-Antipolis >>> 2004 route des lucioles - FR-06902 >>> ------------------------------------------------------------------------ >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject > From fedorov at bwh.harvard.edu Thu Jan 29 16:17:18 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Thu, 29 Jan 2009 16:17:18 -0500 Subject: [slicer-devel] Rigid registration reproducibility In-Reply-To: <82f536810901191226j63ffd262w3787cc9149f97209@mail.gmail.com> References: <82f536810901191226j63ffd262w3787cc9149f97209@mail.gmail.com> Message-ID: <82f536810901291317j3d5bbc14u65c7cb55eda541bd@mail.gmail.com> There was an error with my initial posting. I reported transforms for registering image to itself. But the problem is still there: GUI and CLI results are consistently different with the order of 0.1 difference in translation. Here are the actual transforms for the input test data and parameters I reported earlier, for the sake of completeness. These transforms can be directly loaded into Slicer -- the alignment results are even visibly different. RigidRegistration run from GUI: #Insight Transform File V1.0 # Transform 0 Transform: QuaternionRigidTransform_double_3_3 Parameters: 0.00816444 0.00293686 0.0130089 0.999878 7.54936 5.3547 -0.0257939 FixedParameters: 0.00025 0.00025 0 RigidRegistration run as command line: #Insight Transform File V1.0 # Transform 0 Transform: QuaternionRigidTransform_double_3_3 Parameters: 0.00656795 0.00293988 0.00662475 0.999952 7.76758 5.81825 0.164244 FixedParameters: 0.00025 0.00025 -1.42109e-14 On Mon, Jan 19, 2009 at 3:26 PM, Andriy Fedorov wrote: > Hi, > > I observe that Rigid registration CLI module produces different > (difference in the order of 0.1 for translation) transforms when run > through GUI and using the command line interface. > > I run these on the same machine, and the metric is initialized with > the same seed. The transform is consistent when I repeatedly run it > from command line or GUI, but not between command line and GUI > results. I would expect the same result > > The test images I use are available in the trunk: > $Slicer3/Testing/Data/Input/MRMeningioma0.nrrd (fixed) and > $Slicer3/Testing/Data/Input/MRMeningioma1.nrrd (moving). > > Slicer GUI invocation: > > slicer:0x7f8604605652 --processinformationaddress 0x4734f28 > --fixedsmoothingfactor 0 --movingsmoothingfactor 0 --histogrambins 30 > --spatialsamples 10000 --iterations 100,100,50,20 --learningrate > 0.01,0.005,0.0005,0.0002 --translationscale 100 --outputtransform > /tmp/Slicer3andrey/CHDDI_vtkMRMLLinearTransformNodeE.txt > slicer:0x104e0f0#vtkMRMLScalarVolumeNode1 > slicer:0x104e0f0#vtkMRMLScalarVolumeNode1 > > Transform: > > Baseline transform: QuaternionRigidTransform (0x6dd7c0) > RTTI typeinfo: itk::QuaternionRigidTransform > Reference Count: 4 > Modified Time: 230 > Debug: Off > Observers: > none > Matrix: > 0.999645 -0.0259667 0.00608542 > 0.0260626 0.999529 -0.0162505 > -0.00566058 0.0164033 0.99985 > Offset: [7.54937, 5.35469, -0.0257966] > Center: [0.00025, 0.00025, 0] > Translation: [7.54936, 5.3547, -0.0257939] > Inverse: > 0.999644 0.0260626 -0.00566058 > -0.0259666 0.999528 0.0164033 > 0.00608542 -0.0162505 0.999849 > Singular: 0 > Rotation: 0.00816444 0.00293686 0.0130089 0.999878 > Transform matrix: > 0.999895 -0.0132102 0.0059665 > 0.0132875 0.999826 -0.0130963 > -0.00579246 0.0131742 0.999896 > > Slicer command line invocation: > > ${Slicer3_HOME}/Slicer3 --launch RigidRegistration --iterations > 100,100,50,20 --fixedsmoothingfactor 0 --movingsmoothingfactor 0 > --histogrambins 30 --spatialsamples 10000 --learningrate > 0.01,0.005,0.0005,0.0002 --translationscale 100 > --resampledmovingfilename > /home/andrey/local/src/Slicer/Slicer3-build/Testing/Temporary/RigidRegistrationRegressionTestOutput.nrrd > --outputtransform > /home/andrey/local/src/Slicer/Slicer3-build/Testing/Temporary/RigidRegistrationRegressionTestTransform.txt > /home/andrey/local/src/Slicer/Slicer3/Testing/Data/Input/MRMeningioma0.nrrd > /home/andrey/local/src/Slicer/Slicer3/Testing/Data/Input/MRMeningioma1.nrrd > > Transform: > > Output transform: QuaternionRigidTransform (0x6dda20) > RTTI typeinfo: itk::QuaternionRigidTransform > Reference Count: 4 > Modified Time: 235 > Debug: Off > Observers: > none > Matrix: > 0.999895 -0.0132102 0.0059665 > 0.0132875 0.999826 -0.0130963 > -0.00579246 0.0131742 0.999896 > Offset: [7.76758, 5.81825, 0.164242] > Center: [0.00025, 0.00025, -1.42109e-14] > Translation: [7.76758, 5.81825, 0.164244] > Inverse: > 0.999895 0.0132875 -0.00579246 > -0.0132103 0.999826 0.0131742 > 0.0059665 -0.0130963 0.999897 > Singular: 0 > Rotation: 0.00656795 0.00293988 0.00662475 0.999952 > > -- > Andriy Fedorov > From kikinis at bwh.harvard.edu Fri Jan 30 09:30:11 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis MD) Date: Fri, 30 Jan 2009 09:30:11 -0500 Subject: [slicer-devel] Slicer 3.4 preparations In-Reply-To: <49683630.3040903@bwh.harvard.edu> References: <49683630.3040903@bwh.harvard.edu> Message-ID: <49830EF3.8060002@bwh.harvard.edu> http://wiki.slicer.org/slicerWiki/index.php/Documentation-3.4#List_of_Modules_in_need_of_documentation Hello Slicer Developers, As the code freeze date for Slicer 3.4 moves closer, I would like to point you to the page above where you will have to put documentation for modules which will be distributed as part of the Slicer 3.4 download. You will need to look at Help and Acknowledgements in the module and the reference manual style documentation on the page linked above. You will also have to look at developing tests for your module and proper integration with the mrml tree, where appropriate. Ron -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From vrnova at gmail.com Fri Jan 30 09:45:16 2009 From: vrnova at gmail.com (Yanling Liu) Date: Fri, 30 Jan 2009 09:45:16 -0500 Subject: [slicer-devel] python wrapping in slicer/itk/vtk Message-ID: <24ceacd90901300645j43ec0462o11f73b2d4537bf5a@mail.gmail.com> Hello, could I get some help to configure slicer to be able to interact with slicer, itk, and vtk in the slicer python console? Seems CableSwig is needed to compile itk with python wrapping but CableSwig is not in Slicer-lib folder... Futhermore how can we configure the package to convert between itk and vtk in python? Do we still have *ConnectVTKToITK and* *ConnectITKToVTK *functions to use? Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090130/9f5f2614/attachment.html From marianna at bwh.harvard.edu Fri Jan 30 11:08:27 2009 From: marianna at bwh.harvard.edu (Marianna Jakab) Date: Fri, 30 Jan 2009 11:08:27 -0500 Subject: [slicer-devel] [Fwd: Slicer3 KWMeshVisu] Message-ID: <498325FB.5020507@bwh.harvard.edu> -- Marianna -------------- next part -------------- An embedded message was scrubbed... From: Cristina Gallego Subject: Slicer3 KWMeshVisu Date: Fri, 30 Jan 2009 10:47:57 -0500 Size: 2794 Url: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090130/65a877d6/attachment.eml From vrnova at gmail.com Fri Jan 30 13:44:00 2009 From: vrnova at gmail.com (Yanling Liu) Date: Fri, 30 Jan 2009 13:44:00 -0500 Subject: [slicer-devel] Failed to compile Slicer3 in Windows Vista Message-ID: <24ceacd90901301044u14418b25udf99e67eede280b9@mail.gmail.com> I did a fresh rebuild but got these error messages. Although I am using 64bit system the getbuildtest.tcl generates Win32 projects for Slicer. Is this normal? ------ Build started: Project: EMSegmentBatchMakeModule, Configuration: Release Win32 ------ Linking... bmModuleDescriptionParser.lib(bmModuleParameter.obj) : error LNK2005: "void __cdecl splitString(class std::basic_string,class std::allocator > const &,class std::basic_string,class std::allocator > const &,class std::vector,class std::allocator >,class std::allocator,class std::allocator > > > &)" (?splitString@@YAXABV?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@0AAV?$vector at V?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@V?$allocator at V?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@@2@@2@@Z) already defined in EMSegmentBatchMakeModule.obj Creating library C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\EMSegmentBatchMakeModule.lib and object C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\EMSegmentBatchMakeModule.exp C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\EMSegmentBatchMakeModule.exe : fatal error LNK1169: one or more multiply defined symbols found Build log was saved at "file://c:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\Applications\CLI\BatchMakeApplications\EMSegmentBatchMakeModule\EMSegmentBatchMakeModule.dir\Release\BuildLog.htm" EMSegmentBatchMakeModule - 2 error(s), 0 warning(s) ------ Build started: Project: GaussianBlurBatchMakeModule, Configuration: Release Win32 ------ Linking... bmModuleDescriptionParser.lib(bmModuleParameter.obj) : error LNK2005: "void __cdecl splitString(class std::basic_string,class std::allocator > const &,class std::basic_string,class std::allocator > const &,class std::vector,class std::allocator >,class std::allocator,class std::allocator > > > &)" (?splitString@@YAXABV?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@0AAV?$vector at V?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@V?$allocator at V?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@@2@@2@@Z) already defined in GaussianBlurBatchMakeModule.obj Creating library C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\GaussianBlurBatchMakeModule.lib and object C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\GaussianBlurBatchMakeModule.exp C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\GaussianBlurBatchMakeModule.exe : fatal error LNK1169: one or more multiply defined symbols found Build log was saved at "file://c:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\Applications\CLI\BatchMakeApplications\GaussianBlurBatchMakeModule\GaussianBlurBatchMakeModule.dir\Release\BuildLog.htm" GaussianBlurBatchMakeModule - 2 error(s), 0 warning(s) ------ Build started: Project: RegisterImagesBatchMakeModule, Configuration: Release Win32 ------ Linking... bmModuleDescriptionParser.lib(bmModuleParameter.obj) : error LNK2005: "void __cdecl splitString(class std::basic_string,class std::allocator > const &,class std::basic_string,class std::allocator > const &,class std::vector,class std::allocator >,class std::allocator,class std::allocator > > > &)" (?splitString@@YAXABV?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@0AAV?$vector at V?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@V?$allocator at V?$basic_string at DU ?$char_traits at D@std@@V?$allocator at D@2@@std@@@2@@2@@Z) already defined in RegisterImagesBatchMakeModule.obj Creating library C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\RegisterImagesBatchMakeModule.lib and object C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\RegisterImagesBatchMakeModule.exp C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\RegisterImagesBatchMakeModule.exe : fatal error LNK1169: one or more multiply defined symbols found Build log was saved at "file://c:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\Applications\CLI\BatchMakeApplications\RegisterImagesBatchMakeModule\RegisterImagesBatchMakeModule.dir\Release\BuildLog.htm" RegisterImagesBatchMakeModule - 2 error(s), 0 warning(s) ========== Build: 0 succeeded, 3 failed, 200 up-to-date, 0 skipped ========== -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090130/5f60ef3f/attachment.html From curtislisle at knowledgevis.com Fri Jan 30 13:51:35 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Fri, 30 Jan 2009 13:51:35 -0500 Subject: [slicer-devel] [Fwd: Slicer3 KWMeshVisu] In-Reply-To: <498325FB.5020507@bwh.harvard.edu> References: <498325FB.5020507@bwh.harvard.edu> Message-ID: <8286E8F7-EE60-433E-90EC-9ED989EB1AE5@knowledgevis.com> Hi Marianna, The functionality for displaying, examining, and analyzing mesh elements according to Jacobian, Size, corner angles, etc. for hexahedral meshes does already exist as a Slicer module. This module is called IA_FEMesh, and was added to Slicer in late December. This module is actually a different application than KWMeshVisu, which was developed as a standalone KWWidgets application. The IA_FEMesh module was developed by several faculty members at the University of Iowa. This module can read in surfaces in STL or VTK format, create hexahedral meshes, and analyze and review the meshes. A tutorial has been developed to be included in the Slicer training tutorials. In a separate email, I'll send you training materials for the Slicer module. An overview of the module was presented recently by the developers, and is available here: http://wiki.na-mic.org/Wiki/images/d/d7/NA-MIC_Mesh_Collaboration_2009.ppt Feel free to follow up with me with questions as you are using the module. Curtis Lisle curtislisle at knowledgevis.com On Jan 30, 2009, at 11:08 AM, Marianna Jakab wrote: > > -- > Marianna > > > > > From: Cristina Gallego > Date: January 30, 2009 10:47:57 AM EST > To: marianna at bwh.harvard.edu > Subject: Slicer3 KWMeshVisu > > > Hi Marianna: > > I am wondering if is there any release of the KWMeshVisu module that > i can have access to. I am trying to used some of the SPHARM scripts > for statistical shape modeling and segmentation and it would be > great to have some visualization tools at hand. I understand that > KWMeshVisu was successfully integrated into Slicer3, but i cant find > any release of them online. Any comments or guiding here would be > greatly appreciated. > > Thanks, > > -- > Cristina Gallego > Medical Biophysics M.Sc Candidate > Imaging Research > Sunnybrook Health Sciences Centre > 2075 Bayview Ave, Room S635, Toronto, ON > M4N 3M5 > Tel: (416)-480-6100 x3393 > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090130/7c41c76e/attachment-0001.html From pieper at bwh.harvard.edu Sat Jan 31 15:41:51 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sat, 31 Jan 2009 15:41:51 -0500 Subject: [slicer-devel] python wrapping in slicer/itk/vtk In-Reply-To: <24ceacd90901300645j43ec0462o11f73b2d4537bf5a@mail.gmail.com> References: <24ceacd90901300645j43ec0462o11f73b2d4537bf5a@mail.gmail.com> Message-ID: <4984B78F.6050906@bwh.harvard.edu> Hi Yanling - The python bindings are currently enabled by default for linux and mac builds and are part of the nightly build downloads. It is possible also to build your own python-enabled slicer for windows. The instructions, sample code, and other info is here: http://www.slicer.org/slicerWiki/index.php/Slicer3:Python We're working on getting the nightly build process for windows to include python and hope to have it in place for the 3.4 release. For slicer, "python bindings" means that the script interface is enabled for code that inherits from VTK (which is pretty much all of slicer and the support libraries). So it is possible to manipulate MRML, KWWidgets, and pretty much everything in the program in either C++, Tcl, or Python depending on your specific needs and preferences. Our default python builds include the base python libraries plus the numpy package. Developers may add scipy or other packages to their local builds, but right now we have no good way to package them as part of binary downloads, so slicer modules that depend on them won't be as easy to install for end users. Regarding ITK, I would suggest encapsulating your ITK code either as command line modules or as vtkITK classes depending on which better fits your needs. Either can be invoked from python. There are some example vtkITK classes here that do not use the ConnectVTKToITK-style: http://www.na-mic.org/Wiki/index.php/2009_Winter_Project_Week_vtkITK_Pipeline The direct python wrap of ITK is interesting, but not something we currently support in slicer. My personal feeling is that C++ is the natural language for really using ITK effectively. HTH, -Steve Yanling Liu wrote: > Hello, could I get some help to configure slicer to be able to interact > with slicer, itk, and vtk in the slicer python console? Seems CableSwig > is needed to compile itk with python wrapping but CableSwig is not in > Slicer-lib folder... Futhermore how can we configure the package to > convert between itk and vtk in python? Do we still have > //ConnectVTKToITK and// //ConnectITKToVTK //functions to use? > > Thanks, > Yanling > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Sat Jan 31 16:23:06 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sat, 31 Jan 2009 16:23:06 -0500 Subject: [slicer-devel] Failed to compile Slicer3 in Windows Vista In-Reply-To: <24ceacd90901301044u14418b25udf99e67eede280b9@mail.gmail.com> References: <24ceacd90901301044u14418b25udf99e67eede280b9@mail.gmail.com> Message-ID: <4984C13A.40408@bwh.harvard.edu> Hi Yanling - This is a windows specific bug that cropped up -- the BatchMake team are on it and it should be fixed now. Probably easiest to delete Slicer3-lib/BatchMake* and re-run getbuildtest.tcl. -Steve p.s. I probably should have alerted the list that BatchMake was being added to the default build - sorry for any inconvenience as we get the build and package details sorted out. BatchMake can be used, for example, to run sequences of command line modules are large datasets: http://www.na-mic.org/Wiki/index.php/2009_Winter_Project_Week_ShapePipeline http://batchmake.org/ Yanling Liu wrote: > I did a fresh rebuild but got these error messages. Although I am using > 64bit system the getbuildtest.tcl generates Win32 projects for Slicer. > Is this normal? > > ------ Build started: Project: EMSegmentBatchMakeModule, Configuration: > Release Win32 ------ > Linking... > bmModuleDescriptionParser.lib(bmModuleParameter.obj) : error LNK2005: > "void __cdecl splitString(class std::basic_string std::char_traits,class std::allocator > const &,class > std::basic_string,class > std::allocator > const &,class std::vector std::basic_string,class > std::allocator >,class std::allocator std::basic_string,class > std::allocator > > > &)" > (?splitString@@YAXABV?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@0AAV?$vector at V?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@V?$allocator at V?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@@2@@2@@Z) > already defined in EMSegmentBatchMakeModule.obj > Creating library > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\EMSegmentBatchMakeModule.lib > and object > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\EMSegmentBatchMakeModule.exp > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\EMSegmentBatchMakeModule.exe > : fatal error LNK1169: one or more multiply defined symbols found > Build log was saved at > "file://c:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\Applications\CLI\BatchMakeApplications\EMSegmentBatchMakeModule\EMSegmentBatchMakeModule.dir\Release\BuildLog.htm" > EMSegmentBatchMakeModule - 2 error(s), 0 warning(s) > ------ Build started: Project: GaussianBlurBatchMakeModule, > Configuration: Release Win32 ------ > Linking... > bmModuleDescriptionParser.lib(bmModuleParameter.obj) : error LNK2005: > "void __cdecl splitString(class std::basic_string std::char_traits,class std::allocator > const &,class > std::basic_string,class > std::allocator > const &,class std::vector std::basic_string,class > std::allocator >,class std::allocator std::basic_string,class > std::allocator > > > &)" > (?splitString@@YAXABV?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@0AAV?$vector at V?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@V?$allocator at V?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@@2@@2@@Z) > already defined in GaussianBlurBatchMakeModule.obj > Creating library > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\GaussianBlurBatchMakeModule.lib > and object > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\GaussianBlurBatchMakeModule.exp > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\GaussianBlurBatchMakeModule.exe > : fatal error LNK1169: one or more multiply defined symbols found > Build log was saved at > "file://c:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\Applications\CLI\BatchMakeApplications\GaussianBlurBatchMakeModule\GaussianBlurBatchMakeModule.dir\Release\BuildLog.htm" > GaussianBlurBatchMakeModule - 2 error(s), 0 warning(s) > ------ Build started: Project: RegisterImagesBatchMakeModule, > Configuration: Release Win32 ------ > Linking... > bmModuleDescriptionParser.lib(bmModuleParameter.obj) : error LNK2005: > "void __cdecl splitString(class std::basic_string std::char_traits,class std::allocator > const &,class > std::basic_string,class > std::allocator > const &,class std::vector std::basic_string,class > std::allocator >,class std::allocator std::basic_string,class > std::allocator > > > &)" > (?splitString@@YAXABV?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@0AAV?$vector at V?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@V?$allocator at V?$basic_string at DU?$char_traits at D@std@@V?$allocator at D@2@@std@@@2@@2@@Z) > already defined in RegisterImagesBatchMakeModule.obj > Creating library > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\RegisterImagesBatchMakeModule.lib > and object > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\RegisterImagesBatchMakeModule.exp > C:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\lib\Slicer3\Plugins\Release\RegisterImagesBatchMakeModule.exe > : fatal error LNK1169: one or more multiply defined symbols found > Build log was saved at > "file://c:\cygwin\home\yliu\slicer\Slicer3.3\Slicer3-build\Applications\CLI\BatchMakeApplications\RegisterImagesBatchMakeModule\RegisterImagesBatchMakeModule.dir\Release\BuildLog.htm" > RegisterImagesBatchMakeModule - 2 error(s), 0 warning(s) > ========== Build: 0 succeeded, 3 failed, 200 up-to-date, 0 skipped > ========== > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From fedorov at bwh.harvard.edu Sat Jan 31 16:59:51 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sat, 31 Jan 2009 16:59:51 -0500 Subject: [slicer-devel] Visibility marker in MRML hierarchy Message-ID: <82f536810901311359s72e261d3hbd2f240df9ffad37@mail.gmail.com> Hi, I have a feature suggestion. I think it would be useful to have a visibility marker, which could be toggled within the MRML tree directly (an open/closed "eye" icon next to the MRML node). A similar functionality is available in Paraview, for example. I think this feature would be particularly useful when there is a large number of models in the scene To see what I mean, open PNL brain atlas in Slicer, and try to toggle off visible brain structures. If you are a novice in brain anatomy like me -- this task won't be easy... Andriy Fedorov From fedorov at bwh.harvard.edu Sat Jan 31 19:20:00 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sat, 31 Jan 2009 19:20:00 -0500 Subject: [slicer-devel] Visibility marker in MRML hierarchy In-Reply-To: <4984D7BF.5040906@comcast.net> References: <82f536810901311359s72e261d3hbd2f240df9ffad37@mail.gmail.com> <4984D7BF.5040906@comcast.net> Message-ID: <82f536810901311620x648e582etea184c5fafd39190@mail.gmail.com> With the current interface, one needs to go through the list one by one to see what models are visible. This can be done by mouse right-click at each model and checking visibility flag, or going to Models module and selecting each model one by one. What I suggest is to include visibility in the list, so that by looking at the list you can see what is visible. It's a usability thing, no new functionality. Just a suggestion from user perspective... On Sat, Jan 31, 2009 at 5:59 PM, Alex Yarmarkovich wrote: > Hi Andriy, > > There is a model hierarchy display widget under Models module that does > something similar. Is what you are proposing different from that? > > Alex > > Andriy Fedorov wrote: >> >> Hi, >> >> I have a feature suggestion. >> >> I think it would be useful to have a visibility marker, which could be >> toggled within the MRML tree directly (an open/closed "eye" icon next >> to the MRML node). A similar functionality is available in Paraview, >> for example. >> >> I think this feature would be particularly useful when there is a >> large number of models in the scene To see what I mean, open PNL brain >> atlas in Slicer, and try to toggle off visible brain structures. If >> you are a novice in brain anatomy like me -- this task won't be >> easy... >> >> Andriy Fedorov >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >> > > From luca.antiga at gmail.com Mon Feb 2 04:14:09 2009 From: luca.antiga at gmail.com (Luca Antiga) Date: Mon, 2 Feb 2009 10:14:09 +0100 Subject: [slicer-devel] python wrapping in slicer/itk/vtk In-Reply-To: <4984B78F.6050906@bwh.harvard.edu> References: <24ceacd90901300645j43ec0462o11f73b2d4537bf5a@mail.gmail.com> <4984B78F.6050906@bwh.harvard.edu> Message-ID: <1B110991-DC05-40F1-88E5-EC7FCC0E60A9@gmail.com> Hi guys, sorry for the lag, I've been off the email for a couple of days. I 100% agree with what Steve suggests in his email. However, Python in Slicer also potentially allows to access WrapITK- wrapped ITK (note that I didn't try it first-hand, it just popped into my head). We should enable the following - compile ITK in Slicer3-lib enabling WrapITK (you could also have your own, but this would ensure that only one version of ITK dynamic libraries are loaded - you might get crashes otherwise) (Steve, this is something we could provide upstream) - make the Python modules available to the Slicer3 Python interpreter by setting the proper PYTHONPATH (Steve, same as above) - make Slicer talk to WrapITK through Numpy: on the Slicer side, use image.ToArray() and image.FromArray(), and on the WrapITK side use GetArrayFromImage and GetImageFromArray Something like from Slicer import slicer import itk vol = slicer.MRMLScene.GetNodeFromId('vtkMRMLScalarVolumeNode1') arr = vol.GetImageData().ToArray() itkimage = itk.PyBuffer.IUS2.GetImageFromArray(arr) ... your wrapITK code here ... and outarr = itk.PyBuffer.IUS2.GetArrayFromImage(itkimage) vol.GetImageData().FromArray(outarr) Of course the spacing, origin, direction and stuff should also be set, but we could provide some Python code in the Slicer module to make the connection. What do you guys think? Luca On Jan 31, 2009, at 9:41 PM, Steve Pieper wrote: > Hi Yanling - > > The python bindings are currently enabled by default for linux and mac > builds and are part of the nightly build downloads. It is possible > also > to build your own python-enabled slicer for windows. The > instructions, > sample code, and other info is here: > > http://www.slicer.org/slicerWiki/index.php/Slicer3:Python > > We're working on getting the nightly build process for windows to > include python and hope to have it in place for the 3.4 release. > > For slicer, "python bindings" means that the script interface is > enabled > for code that inherits from VTK (which is pretty much all of slicer > and > the support libraries). So it is possible to manipulate MRML, > KWWidgets, and pretty much everything in the program in either C++, > Tcl, > or Python depending on your specific needs and preferences. > > Our default python builds include the base python libraries plus the > numpy package. Developers may add scipy or other packages to their > local builds, but right now we have no good way to package them as > part > of binary downloads, so slicer modules that depend on them won't be as > easy to install for end users. > > Regarding ITK, I would suggest encapsulating your ITK code either as > command line modules or as vtkITK classes depending on which better > fits > your needs. Either can be invoked from python. There are some > example > vtkITK classes here that do not use the ConnectVTKToITK-style: > > http://www.na-mic.org/Wiki/index.php/2009_Winter_Project_Week_vtkITK_Pipeline > > The direct python wrap of ITK is interesting, but not something we > currently support in slicer. My personal feeling is that C++ is the > natural language for really using ITK effectively. > > HTH, > -Steve > > Yanling Liu wrote: >> Hello, could I get some help to configure slicer to be able to >> interact >> with slicer, itk, and vtk in the slicer python console? Seems >> CableSwig >> is needed to compile itk with python wrapping but CableSwig is not in >> Slicer-lib folder... Futhermore how can we configure the package to >> convert between itk and vtk in python? Do we still have >> //ConnectVTKToITK and// //ConnectITKToVTK //functions to use? >> >> Thanks, >> Yanling >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject -- Luca Antiga, PhD Head, Medical Imaging Unit, Biomedical Engineering Department, Mario Negri Institute. mail: Villa Camozzi, 24020, Ranica (BG), Italy phone: +39 035 4535-381 email: antiga at marionegri.it web: http://villacamozzi.marionegri.it/~luca -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090202/5a8ca586/attachment-0001.html From jvs at bwh.harvard.edu Mon Feb 2 08:54:00 2009 From: jvs at bwh.harvard.edu (Julien von Siebenthal) Date: Mon, 02 Feb 2009 08:54:00 -0500 Subject: [slicer-devel] python wrapping in slicer/itk/vtk In-Reply-To: <1B110991-DC05-40F1-88E5-EC7FCC0E60A9@gmail.com> References: <24ceacd90901300645j43ec0462o11f73b2d4537bf5a@mail.gmail.com> <4984B78F.6050906@bwh.harvard.edu> <1B110991-DC05-40F1-88E5-EC7FCC0E60A9@gmail.com> Message-ID: <4986FAF8.3020202@bwh.harvard.edu> Luca, Totally in favor of WrapITK. I think it is worth to give it a try. I myself used it. It is integrated to the itk dist and the changes to cmake are not that difficult. I integrated the rician filter of Slicer part of an itk pipeline written in python. I can provide some help with example of my changes if needed. I am not an expert of cmake, thus I assume quite a lot of improvements can be achieved. Additionally, there is an issue with numpy/itk conversion: it works, but sometimes generate exceptions related to multithreading, possibly critical sections of the numpy conversion code needs attention. Best, Julien -- Julien von Siebenthal, Ph.D. Psychiatric Neuroimaging Laboratory, Harvard Medical School Brigham and Women's Hospital, 75 Francis St., Boston MA 02115 jvs at bwh.harvard.edu Luca Antiga wrote: > Hi guys, > sorry for the lag, I've been off the email for a couple of days. > I 100% agree with what Steve suggests in his email. > > However, Python in Slicer also potentially allows to access > WrapITK-wrapped > ITK (note that I didn't try it first-hand, it just popped into my head). > > We should enable the following > - compile ITK in Slicer3-lib enabling WrapITK (you could also have > your own, > but this would ensure that only one version of ITK dynamic libraries > are loaded - > you might get crashes otherwise) (Steve, this is something we could > provide upstream) > - make the Python modules available to the Slicer3 Python interpreter > by setting > the proper PYTHONPATH (Steve, same as above) > - make Slicer talk to WrapITK through Numpy: on the Slicer side, use > image.ToArray() > and image.FromArray(), and on the WrapITK side use GetArrayFromImage > and GetImageFromArray > > Something like > > from Slicer import slicer > import itk > > vol = slicer.MRMLScene.GetNodeFromId('vtkMRMLScalarVolumeNode1') > arr = vol.GetImageData().ToArray() > itkimage = itk.PyBuffer.IUS2.GetImageFromArray(arr) > > ... your wrapITK code here ... > > and > > outarr = itk.PyBuffer.IUS2.GetArrayFromImage(itkimage) > vol.GetImageData().FromArray(outarr) > > Of course the spacing, origin, direction and stuff should also be set, > but we could provide some > Python code in the Slicer module to make the connection. > > What do you guys think? > > > Luca > > > On Jan 31, 2009, at 9:41 PM, Steve Pieper wrote: > >> Hi Yanling - >> >> The python bindings are currently enabled by default for linux and mac >> builds and are part of the nightly build downloads. It is possible also >> to build your own python-enabled slicer for windows. The instructions, >> sample code, and other info is here: >> >> http://www.slicer.org/slicerWiki/index.php/Slicer3:Python >> >> We're working on getting the nightly build process for windows to >> include python and hope to have it in place for the 3.4 release. >> >> For slicer, "python bindings" means that the script interface is enabled >> for code that inherits from VTK (which is pretty much all of slicer and >> the support libraries). So it is possible to manipulate MRML, >> KWWidgets, and pretty much everything in the program in either C++, Tcl, >> or Python depending on your specific needs and preferences. >> >> Our default python builds include the base python libraries plus the >> numpy package. Developers may add scipy or other packages to their >> local builds, but right now we have no good way to package them as part >> of binary downloads, so slicer modules that depend on them won't be as >> easy to install for end users. >> >> Regarding ITK, I would suggest encapsulating your ITK code either as >> command line modules or as vtkITK classes depending on which better fits >> your needs. Either can be invoked from python. There are some example >> vtkITK classes here that do not use the ConnectVTKToITK-style: >> >> http://www.na-mic.org/Wiki/index.php/2009_Winter_Project_Week_vtkITK_Pipeline >> >> The direct python wrap of ITK is interesting, but not something we >> currently support in slicer. My personal feeling is that C++ is the >> natural language for really using ITK effectively. >> >> HTH, >> -Steve >> >> Yanling Liu wrote: >>> Hello, could I get some help to configure slicer to be able to interact >>> with slicer, itk, and vtk in the slicer python console? Seems CableSwig >>> is needed to compile itk with python wrapping but CableSwig is not in >>> Slicer-lib folder... Futhermore how can we configure the package to >>> convert between itk and vtk in python? Do we still have >>> //ConnectVTKToITK and// //ConnectITKToVTK //functions to use? >>> >>> Thanks, >>> Yanling >>> >>> >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as >>> the subject >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > > -- > Luca Antiga, PhD > Head, Medical Imaging Unit, > Biomedical Engineering Department, > Mario Negri Institute. > mail: Villa Camozzi, 24020, Ranica (BG), Italy > phone: +39 035 4535-381 > email: antiga at marionegri.it > web: http://villacamozzi.marionegri.it/~luca > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Mon Feb 2 09:21:54 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 02 Feb 2009 09:21:54 -0500 Subject: [slicer-devel] [slicer-users] probe volume with model In-Reply-To: <4985BBDF.7010204@med.unideb.hu> References: <4985BBDF.7010204@med.unideb.hu> Message-ID: <49870182.3060006@bwh.harvard.edu> Hi Andras - Looks like a bug crept into the scalar display code - I hope this checkin fixes it: http://viewvc.slicer.org/viewcvs.cgi/trunk/Base/GUI/vtkSlicerViewerWidget.cxx?rev=8493&view=diff&r1=8493&r2=8492&p1=trunk/Base/GUI/vtkSlicerViewerWidget.cxx&p2=/trunk/Base/GUI/vtkSlicerViewerWidget.cxx It would be good for people to check that this doesn't break something else. Regarding the stochastic tractography, Julien von Siebenthal is the best person to contact - basically he's reimplemented it in python and that's were the active development is. Thanks for the report! Steve Andras Jakab wrote: > Hello, > > I was just wondering if the Probe volume with model uses any GPU > features that my macbook lacks, or am I doing something wrong? > Apparently I use slicer2 generated scalars exported in nrrd (like FA, > trace) and would like to get a simplified brain surface painted > according to trace values of the underlying voxels. It turns out to be > gray regardless of the palette or any other settings change. In the > models tab should the "show scalars" box be checked in? Well, tried many > options. > > Thanks a lot! > ... and the new 3.3 codes seem to have many interesting features, I'll > test 'em both on linux64 and osx... (is the python stochastic > tratography different from the "old" one in any ways?) > > Cheers, > Andras Jakab > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From nicole at bwh.harvard.edu Mon Feb 2 14:00:46 2009 From: nicole at bwh.harvard.edu (Nicole Aucoin) Date: Mon, 2 Feb 2009 14:00:46 -0500 (EST) Subject: [slicer-devel] Meeting tomorrow, Feb 3/09, 3pm Message-ID: Slicer developer's meeting, Feb 3 at 3pm, second floor demo room, 1249 Boylston St., Boston. Agenda: informatics and remote data handling in preparation for the mBIRN meeting, (code freeze). Nicole -- Nicole Aucoin nicole at bwh.harvard.edu Surgical Planning Laboratory, Brigham and Women's Hospital Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 http://www.slicer.org From inorton at partners.org Mon Feb 2 16:21:30 2009 From: inorton at partners.org (Isaiah Norton) Date: Mon, 02 Feb 2009 16:21:30 -0500 Subject: [slicer-devel] Visibility marker in MRML hierarchy In-Reply-To: <1233609263.16740.32.camel@thanks.bwh.harvard.edu> References: <82f536810901311359s72e261d3hbd2f240df9ffad37@mail.gmail.com> <4984D7BF.5040906@comcast.net> <82f536810901311620x648e582etea184c5fafd39190@mail.gmail.com> <1233609263.16740.32.camel@thanks.bwh.harvard.edu> Message-ID: <1233609690.16740.37.camel@thanks.bwh.harvard.edu> A few suggestions along this line: For the Model hierarchy: 1) make everything in the list deletable. Right now, only model hierarchy nodes are deletable. Actual models must be deleted through the main MRML hierarchy. For the main MRML hierarchy: 2) little icons for different datatypes in the hierarchy. Maybe combined with visibility control (for models). That is: the datatype icon could be shaded when invisible. This would be a very helpful visual cue - in line with the plethora of other little icons :) Though probably would need an upstream enhancement to the widget. In the interim, maybe there could be a middle-ground between current option to display or hide the full node classname (eg vtkLinearTransformNode). For example a short parenthetical tag at the end of the identifier, like "SPGR GAD (VOL)" "EMSeg Cortex (MOD)" etc.. 3) Add function to sort by data type (but preserve the groupings correctly on return!) 4) Add hierarchy grouping similar to what is available in model list. Yes, it is possible to group inside identity transform, but this is not optimal (one reason: http://na-mic.org/Mantis/view.php?id=404 ) Maybe these belong in a usability discussion somewhere on the wiki.. -I > On Sat, 2009-01-31 at 19:20 -0500, Andriy Fedorov wrote: > > With the current interface, one needs to go through the list one by > > one to see what models are visible. This can be done by mouse > > right-click at each model and checking visibility flag, or going to > > Models module and selecting each model one by one. > > > > What I suggest is to include visibility in the list, so that by > > looking at the list you can see what is visible.be very helpful. > > > > It's a usability thing, no new functionality. Just a suggestion from > > user perspective... > > > > > > On Sat, Jan 31, 2009 at 5:59 PM, Alex Yarmarkovich wrote: > > > Hi Andriy, > > > > > > There is a model hierarchy display widget under Models module that does > > > something similar. Is what you are proposing different from that? > > > > > > Alex > > > > > > Andriy Fedorov wrote: > > >> > > >> Hi, > > >> > > >> I have a feature suggestion. > > >> > > >> I think it would be useful to have a visibility marker, which could be > > >> toggled within the MRML tree directly (an open/closed "eye" icon next > > >> to the MRML node). A similar functionality is available in Paraview, > > >> for example. > > >> > > >> I think this feature would be particularly useful when there is a > > >> large number of models in the scene To see what I mean, open PNL brain > > >> atlas in Slicer, and try to toggle off visible brain structures. If > > >> you are a novice in brain anatomy like me -- this task won't be > > >> easy... > > >> > > >> Andriy Fedorov > > >> _______________________________________________ > > >> slicer-devel mailing list > > >> slicer-devel at bwh.harvard.edu > > >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > >> To unsubscribe: send email to > > >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > > >> subject > > >> > > > > > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From pieper at bwh.harvard.edu Tue Feb 3 17:43:31 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 03 Feb 2009 17:43:31 -0500 Subject: [slicer-devel] follow up points: Re: [NAMIC-All] Slicer 3.4 preparations In-Reply-To: <49830EF3.8060002@bwh.harvard.edu> References: <49683630.3040903@bwh.harvard.edu> <49830EF3.8060002@bwh.harvard.edu> Message-ID: <4988C893.4040202@bwh.harvard.edu> Hi - A few follow up points on the slicer 3.4 release planning and what it means for developers: * If you have any new functionality that should be included in the 3.4 release, please remember that you must check it in by the 5pm EST *tomorrow* so it can be incorporated in the builds and testing. * We will use the month of February to debug and tune the core set of functionality, and develop documentation for base code and modules. *Debugging; not new features*. Please focus on fixing: Crashes, Performance, Usability, and Leaks. * During the month of February we will freeze adding any new features to the svn trunk. We will rely on the honor system for this, but remember there is a public log of everything you do so you will not be able to cheat! * By the end of February we will start creating Release Candidate builds which will then be tested during March. * When we get close to the final 3.4 release (probably by end of March) we will create a release branch and re-open the trunk for development of the next version. * Please review the bug tracker: look for any issues that are 'yours' or that you can fix, even if they are assigned to someone else. Add notes if you intend to work on the bug or if you think it should not be included in the 3.4 release. http://www.na-mic.org/Bug/my_view_page.php * If you want to work on new features and/or infrastructure changes to slicer3 during February or March, you must create your own experimental branch in the repository. You can follow the instructions here to create your branch: http://www.na-mic.org/Wiki/index.php/Engineering:Subversion_Repository After the release branch is created, developers will be able to merge changes from their branches into the trunk of slicer3 for the next release. Any questions, please contact me or Ron. Thanks, Steve Ron Kikinis MD wrote: > http://wiki.slicer.org/slicerWiki/index.php/Documentation-3.4#List_of_Modules_in_need_of_documentation > > > Hello Slicer Developers, > > As the code freeze date for Slicer 3.4 moves closer, I would like to > point you to the page above where you will have to put documentation for > modules which will be distributed as part of the Slicer 3.4 download. > > You will need to look at Help and Acknowledgements in the module and the > reference manual style documentation on the page linked above. You will > also have to look at developing tests for your module and proper > integration with the mrml tree, where appropriate. > > Ron > From Stephen.Aylward at Kitware.com Wed Feb 4 09:58:18 2009 From: Stephen.Aylward at Kitware.com (Stephen Aylward) Date: Wed, 4 Feb 2009 09:58:18 -0500 Subject: [slicer-devel] Teem -vs- TEEM -vs Teem Message-ID: Hi, Two bugs 1) Teem CMakeLists.txt, line 70, declares TEEM_VTK_TOOLKITS_IPATH as a bool, when it should be a FILEPATH. If you run cmake in the build dir it requires tom-foolery to get teem to compile as needed for Slicer. 2) The svn command in getbuildtest2/genlib2 checkouts a version of Teem that builds the project TEEM. However, the cmakelists.txt file in Slicer3, line 155, does a find_package(Teem REQUIRED) instead of find_package(TEEM). Probably easiest to fix in Teem checkout, but config files will need to be renamed, etc. Caps matter on Linux - because find_package won't be able to find TEEMConfig.cmake or TEEMUse.cmake in the build dir. Stephen -- Stephen R. Aylward, Ph.D. Chief Medical Scientist Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090204/d948ea68/attachment.html From pieper at bwh.harvard.edu Wed Feb 4 10:06:19 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 04 Feb 2009 10:06:19 -0500 Subject: [slicer-devel] Teem -vs- TEEM -vs Teem In-Reply-To: References: Message-ID: <4989AEEB.801@bwh.harvard.edu> Hi Stephen - Bill Lorensen noticed issue 1 in Salt Lake, so David Cole is to take a look at the Teem cmake files to clean them up. For now I guess you need to workaround with args on the command line or edit the cache. Regarding issue 2, getbuildtest2 is still considered 'experimental' but in fact it probably should be deprecated and removed since it isn't being maintained. It was created as a workaround for people who couldn't access cvs due to firewalls. (Perhaps still needed for people in GE India, Jim?). Gordon changed from TEEM to Teem for his 1.10 release so we reworked getbuildtest to use that notation consistently. Thanks, Steve Stephen Aylward wrote: > Hi, > > Two bugs > > 1) Teem CMakeLists.txt, line 70, declares TEEM_VTK_TOOLKITS_IPATH as a > bool, when it should be a FILEPATH. If you run cmake in the build dir > it requires tom-foolery to get teem to compile as needed for Slicer. > > 2) The svn command in getbuildtest2/genlib2 checkouts a version of Teem > that builds the project TEEM. However, the cmakelists.txt file in > Slicer3, line 155, does a find_package(Teem REQUIRED) instead of > find_package(TEEM). Probably easiest to fix in Teem checkout, but > config files will need to be renamed, etc. Caps matter on Linux - > because find_package won't be able to find TEEMConfig.cmake or > TEEMUse.cmake in the build dir. > > > Stephen > > > > -- > Stephen R. Aylward, Ph.D. > Chief Medical Scientist > Kitware, Inc. - North Carolina Office > http://www.kitware.com > (518) 371-3971 x300 > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From fedorov at bwh.harvard.edu Wed Feb 4 19:47:56 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Wed, 4 Feb 2009 19:47:56 -0500 Subject: [slicer-devel] Slicer build on Mac: cmCPackDragNDropGenerator.h not found Message-ID: <82f536810902041647v66fdfe95g669de68ad050789e@mail.gmail.com> Hi, I have error building Slicer during CMake compilation on Mac: Slicer3-lib/CMake/Source/CPack/cmCPackGeneratorFactory.cxx:28:41: error: cmCPackDragNDropGenerator.h: No such file or directory This seems strange, because I know many people use Mac's, and I haven't seen posts about this problem recently. Am I missing something simple, or is this a bug? Andriy Fedorov From fedorov at bwh.harvard.edu Thu Feb 5 09:30:07 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Thu, 5 Feb 2009 09:30:07 -0500 Subject: [slicer-devel] Slicer build on Mac: cmCPackDragNDropGenerator.h not found In-Reply-To: <82f536810902041647v66fdfe95g669de68ad050789e@mail.gmail.com> References: <82f536810902041647v66fdfe95g669de68ad050789e@mail.gmail.com> Message-ID: <82f536810902050630n1e89d4e7me37f99d5ed33098f@mail.gmail.com> I see the missing files have been added to the CMake branch as of today. On Wed, Feb 4, 2009 at 7:47 PM, Andriy Fedorov wrote: > Hi, > > I have error building Slicer during CMake compilation on Mac: > > Slicer3-lib/CMake/Source/CPack/cmCPackGeneratorFactory.cxx:28:41: > error: cmCPackDragNDropGenerator.h: No such file or directory > > This seems strange, because I know many people use Mac's, and I > haven't seen posts about this problem recently. > > Am I missing something simple, or is this a bug? > > Andriy Fedorov > From pieper at bwh.harvard.edu Thu Feb 5 14:39:57 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 05 Feb 2009 14:39:57 -0500 Subject: [slicer-devel] slicer 3.4 usage testing session Message-ID: <498B408D.1000205@bwh.harvard.edu> Hi - Ron and I are planning to spend the afternoon of Tues, Feb 24 going through a variety of activities (putting ourselves in the shoes of the slicer 3.4 users). Other developers are welcome to attend so we can try out running the application end-to-end for some common tasks to make a list of required fixes. http://www.slicer.org/slicerWiki/index.php/Slicer3:Using3.4 -Steve From pieper at bwh.harvard.edu Thu Feb 5 17:52:49 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 05 Feb 2009 17:52:49 -0500 Subject: [slicer-devel] [slicer-users] probe volume with model In-Reply-To: <498B18DF.3060808@dote.hu> References: <4985BBDF.7010204@med.unideb.hu> <49870182.3060006@bwh.harvard.edu> <498B18DF.3060808@dote.hu> Message-ID: <498B6DC1.50507@bwh.harvard.edu> Hi Andras - We actually have a bug report in about the image rendering - there's a snippet of tcl code in the bug report that can be used as a rendering workaround for now: http://www.na-mic.org/Bug/view.php?id=259 I'm hoping to hook this to a menu for the 3.4 release. Let us know how it goes with the scalar visibility. I believe now it is doing an auto-range calculation. It may get thrown off if there are spurious values. Regarding the average DWI gradient: that would probably be a quick python module - any python hackers willing to take up the challenge? Either that, or there may also be a one-line unu command to do it (check http://teem.sf.net if you want to try that path). Otherwise I guess it's a feature request for the bug tracker. -Steve p.s. beautiful artwork! I'm inspired! Andras Jakab wrote: > Hello, > > Thanks a lot! > > Looks like the 2009-02-03 fixes the problem, however, I noticed that > it's hard to find a palette for the scalar map I'm using. (most of the > color LUTs are not suitable, only one end of the spectrum is used, > perhaps it's due to the fact that the voxel values range from 2-8... > I'll try "intensity" normalization, hope it helps) > > One more request / report: Is it possible to generate the average > gradient DWI volume in slicer 3? (and export it as .nrrd) It would be > really helpful for us to perform a DTI-to-other-modality registration, > like using a tcl interactor code or any solution to enable this feature > in slicer3 as well... > > Regards, > Andr?s Jakab > > p.s., offtopic: the "manual" export large image feature is really > useful... Slicer's feature as an artistic tool was exploited to render > this: http://aokhok.unideb.hu/radiology/allfibers_artistic.jpg and I had > it printed in A2 poster size to decorate the workplace. > If I were a programmer, I would code a batch command which rotates the > camera angle in discrete steps and saves the images for the video needs. > > Steve Pieper wrote: >> Hi Andras - >> >> Looks like a bug crept into the scalar display code - I hope this >> checkin fixes it: >> >> http://viewvc.slicer.org/viewcvs.cgi/trunk/Base/GUI/vtkSlicerViewerWidget.cxx?rev=8493&view=diff&r1=8493&r2=8492&p1=trunk/Base/GUI/vtkSlicerViewerWidget.cxx&p2=/trunk/Base/GUI/vtkSlicerViewerWidget.cxx >> >> It would be good for people to check that this doesn't break something else. >> >> Regarding the stochastic tractography, Julien von Siebenthal is the best >> person to contact - basically he's reimplemented it in python and that's >> were the active development is. >> >> Thanks for the report! >> Steve >> >> >> Andras Jakab wrote: >> >>> Hello, >>> >>> I was just wondering if the Probe volume with model uses any GPU >>> features that my macbook lacks, or am I doing something wrong? >>> Apparently I use slicer2 generated scalars exported in nrrd (like FA, >>> trace) and would like to get a simplified brain surface painted >>> according to trace values of the underlying voxels. It turns out to be >>> gray regardless of the palette or any other settings change. In the >>> models tab should the "show scalars" box be checked in? Well, tried many >>> options. >>> >>> Thanks a lot! >>> ... and the new 3.3 codes seem to have many interesting features, I'll >>> test 'em both on linux64 and osx... (is the python stochastic >>> tratography different from the "old" one in any ways?) >>> >>> Cheers, >>> Andras Jakab >>> _______________________________________________ >>> slicer-users mailing list >>> slicer-users at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >>> To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >> _______________________________________________ >> slicer-users mailing list >> slicer-users at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >> To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject >> >> > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From Stephen.Aylward at Kitware.com Fri Feb 6 10:21:18 2009 From: Stephen.Aylward at Kitware.com (Stephen Aylward) Date: Fri, 6 Feb 2009 10:21:18 -0500 Subject: [slicer-devel] tcl8.5 Message-ID: Hi, Tried enabling tcl8.5 via the option in getbuildtest, and I get the following error: *** No rule to make target `/home/aylward/src/DashboardSrcs/Slicer3/../Slicer3-lib/tcl-build/lib/ libtcl8.4.so', needed by `bin/libvtkCommonTCL.so.5.3.0'. Stop. I'll go back to using 8.4 s -- Stephen R. Aylward, Ph.D. Chief Medical Scientist Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090206/38d944f2/attachment.html From pieper at bwh.harvard.edu Fri Feb 6 11:08:53 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 06 Feb 2009 11:08:53 -0500 Subject: [slicer-devel] tcl8.5 In-Reply-To: References: Message-ID: <498C6095.3040903@bwh.harvard.edu> Hi Stephen - The upgrade to tcl8.5 got stalled due to a problem building BLT on the mac. So I guess we'll stick with 8.4 for slicer3.4. Attila is looking at the cvs head of BLT (and new incrTcl/iwidgets) for his solaris builds so that might also work for the mac and we could upgrade across the board after the release branch. -Steve Stephen Aylward wrote: > Hi, > > Tried enabling tcl8.5 via the option in getbuildtest, and I get the > following error: > > *** No rule to make target > `/home/aylward/src/DashboardSrcs/Slicer3/../Slicer3-lib/tcl-build/lib/libtcl8.4.so > ', needed by `bin/libvtkCommonTCL.so.5.3.0'. Stop. > > > I'll go back to using 8.4 > > s > > -- > Stephen R. Aylward, Ph.D. > Chief Medical Scientist > Kitware, Inc. - North Carolina Office > http://www.kitware.com > (518) 371-3971 x300 > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From kikinis at bwh.harvard.edu Fri Feb 6 14:45:38 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Fri, 06 Feb 2009 14:45:38 -0500 Subject: [slicer-devel] Action Item: Documentation-3.4 Message-ID: <498C9362.90000@bwh.harvard.edu> http://www.slicer.org/slicerWiki/index.php/Documentation-3.4 Hello Slicer Developers, after the feature freeze for Slicer 3.4, I have generated a listing of all Modules contained in the nightly build. We have added names for each module as to who we think the owner of that module is. Please do the following by Tuesday 2/10: -Review the list and see if your modules are listed. If you want them removed, please do so. If I missed yours, and it should be listed, let me know. -The modules are organized hierarchically. Some of the modules are at odd places for historic reasons. Please move your module where it belongs. -Please send me an email to acknowledge that you have looked at this. And remember: If your module is listed, we will expect you to provide documentation and other things as listed at the top of the page which is linked above. Ron -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From hata at bwh.harvard.edu Sat Feb 7 16:01:27 2009 From: hata at bwh.harvard.edu (Nobuhiko Hata) Date: Sat, 07 Feb 2009 16:01:27 -0500 Subject: [slicer-devel] NDI Demonstrates Support of IGSTK and 3D Slicer Software for Medical Research In-Reply-To: <48289566.5090408@kitware.com> References: <48289566.5090408@kitware.com> Message-ID: <498DF6A7.30506@bwh.harvard.edu> Dear NA-MIC community As a part of Image Guided Therapy effort in NA-MIC, we collaborated with Northern Digital Inc to develop therir product demo USING Slicer and IGSTK. Today, they annouced its release at IGSTK's users meeting. I was told that NDI will start distributing the demonstration with Slicer at exhibitions. Please see their product annoucement I attached to this email. The annoucement can be also found at NDI's website. http://www.ndigital.com/medical/pr-igstk3dslicer-feb2009.php Please witness the power of open source and impact of NA-MIC there again. Nobuhiko Hata, PhD Brigham and Women's Hospital www.snrlab.org ======= NDI (Northern Digital Inc.), a world-leader in 3D measurement technology for medical, industrial and research applications, is pleased to announce they will be demonstrating how open source software can be used to quickly build research applications that integrate NDI technology at the first ever IGSTK User Group Meeting. An example of an image-guided needle targeting application using NDI?s Aurora System will be shown at the meeting taking place in Orlando, Florida, on February 7, 2009 and being held in conjunction with the 2009 SPIE Medical Imaging Conference. An 18 gauge needle containing a tiny electromagnetic sensor at its tip will be used to demonstrate how the needle can be displayed on a diagnostic image, and updated in real-time as it is manipulated to accurately target a lesion represented in an anatomical phantom. The software component of the demonstration was built using two open source packages specifically designed to support medical imaging and image-guided surgery research. Jeff Stanley, Technical Account Manager for Medical Research at NDI commented, "NDI continues to value our long-standing relationships with medical researchers in the academic community. We are pleased to show our support of initiatives designed to help researchers more efficiently develop software applications that integrate our tracking technologies. This work will allow new users and existing users to focus their energies on answering their research questions, rather than using valuable resources to develop the tools to get them there." The first package used was the Image-Guided Surgery Toolkit (IGSTK), a project led by the Department of Radiology?s Imaging Science and Information Systems (ISIS) Center at Georgetown University Medical Center in Washington. IGSTK is an open source software toolkit, which provides the basic software components needed to quickly develop image-guided surgery applications. This project is a collaboration between Georgetown University, Kitware, Inc., Arizona State University, and SINTEF from Norway. The principal investigator of the IGSTK project is Dr. Kevin Cleary, Director of the ISIS Center. The second package used was 3D Slicer, open source software for visualization, registration, segmentation, and quantification of medical data. This development effort has been enabled by the participation of several large scale NIH funded efforts. Principal investigators on this project include Dr. Ron Kikinis from the National Alliance of Medical Image Computing (NA-MIC), and Dr. Ferenc Jolesz from the National Center for Image Guided Therapy (NCIGT), both based out of the Brigham & Women?s Hospital in Boston. Other communities involved in the development of the code include the NAC, BIRN and CIMIT research organizations. Support for both of these open source software initiatives comes from several federal funding sources including the National Institutes of Health (NIH) and the National Institute of Biomedical Imaging and Bioengineering (NIBIB) as well as many others. For more information about these projects please visit http://www.igstk.org and http://www.slicer.org. For more information about NDI and their products, please visit http://www.ndigital.com/medical/. About NDI Established over 25 years ago, NDI is trusted by international leaders in medicine, industry and research for the accuracy and reliability of its measurement technology. NDI systems are used in applications from computer-assisted surgery to aeronautics; from quality inspection to human motion research. Today, the company is a world leader in advanced 3D measurement technology. With over 14,000 installations in more than 30 countries around the world, NDI?s success has been built on the ability to understand and meet the unique individual needs of customer applications. From bill.lorensen at gmail.com Sat Feb 7 16:11:23 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sat, 7 Feb 2009 16:11:23 -0500 Subject: [slicer-devel] Execution Model Documentation Message-ID: <4db4735c0902071311u22bc699rfbdbcbe190f9c531@mail.gmail.com> Folks, I have checked in changes to vtkCommandLineModuleGUI that incorporate the information into the Help section of execution model commands. According to the documentation standards outlined in: http://wiki.slicer.org/slicerWiki/index.php/Documentation-3.4#Other_Modules "Please add a pointer to the documentation on the Slicer wiki to the the Help tab of the module. See the Editor module in Slicer for an example." Now, if you provide a , a link to that url will appear in the Help section of the GUI for your module. For example, the Filtering->Arithmetic->Add filter has the following http://www.slicer.org/slicerWiki/index.php/Modules:AddImages-Documentation-3.4 In the future you can expect additional facilities that will leverage your execution model documentation. Bill From bill.lorensen at gmail.com Sun Feb 8 08:27:21 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sun, 8 Feb 2009 08:27:21 -0500 Subject: [slicer-devel] Bad Link to SVN in Documentation Template Message-ID: <4db4735c0902080527l75f55ea0y42a2c7aad577e23d@mail.gmail.com> Guys, I think the link to the source repository in the documentation template: http://wiki.slicer.org/slicerWiki/index.php/Slicer3:Module_Documentation-3.4_Template is wrong. I believe that http://www.na-mic.org/ViewVC/index.cgi/ points to the old Slicer3 repository? What is the correct link for ViewVC? Bill From fedorov at bwh.harvard.edu Sun Feb 8 09:26:24 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sun, 8 Feb 2009 09:26:24 -0500 Subject: [slicer-devel] Bad Link to SVN in Documentation Template In-Reply-To: <4db4735c0902080527l75f55ea0y42a2c7aad577e23d@mail.gmail.com> References: <4db4735c0902080527l75f55ea0y42a2c7aad577e23d@mail.gmail.com> Message-ID: <82f536810902080626h641269c9v1245e5c66fff3438@mail.gmail.com> Bill, I think the correct link is http://viewvc.slicer.org/ The documentation link http://www.na-mic.org/Slicer/Documentation/Slicer3/html/ is also incorrect (out of date). Fedorov On Sun, Feb 8, 2009 at 8:27 AM, Bill Lorensen wrote: > Guys, > > I think the link to the source repository in the documentation template: > http://wiki.slicer.org/slicerWiki/index.php/Slicer3:Module_Documentation-3.4_Template > > is wrong. I believe that > http://www.na-mic.org/ViewVC/index.cgi/ > > points to the old Slicer3 repository? > > What is the correct link for ViewVC? > > Bill > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From bill.lorensen at gmail.com Sun Feb 8 12:46:24 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sun, 8 Feb 2009 12:46:24 -0500 Subject: [slicer-devel] Bad Link to SVN in Documentation Template In-Reply-To: <82f536810902080626h641269c9v1245e5c66fff3438@mail.gmail.com> References: <4db4735c0902080527l75f55ea0y42a2c7aad577e23d@mail.gmail.com> <82f536810902080626h641269c9v1245e5c66fff3438@mail.gmail.com> Message-ID: <4db4735c0902080946k3c5f53bbg82ffe9cdb8141c4b@mail.gmail.com> Andriy, Thanks. That new link uses ViewCVS while the odl one used ViewVC. I wonder why it changed to ViewCVS. Seems like a step backward. Bill On Sun, Feb 8, 2009 at 9:26 AM, Andriy Fedorov wrote: > Bill, > > I think the correct link is http://viewvc.slicer.org/ > > The documentation link > http://www.na-mic.org/Slicer/Documentation/Slicer3/html/ is also > incorrect (out of date). > > Fedorov > > > > On Sun, Feb 8, 2009 at 8:27 AM, Bill Lorensen wrote: >> Guys, >> >> I think the link to the source repository in the documentation template: >> http://wiki.slicer.org/slicerWiki/index.php/Slicer3:Module_Documentation-3.4_Template >> >> is wrong. I believe that >> http://www.na-mic.org/ViewVC/index.cgi/ >> >> points to the old Slicer3 repository? >> >> What is the correct link for ViewVC? >> >> Bill >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > From nagy.attila at yahoo.com Mon Feb 9 03:04:04 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Mon, 9 Feb 2009 00:04:04 -0800 (PST) Subject: [slicer-devel] tcl8.5 In-Reply-To: <498C6095.3040903@bwh.harvard.edu> Message-ID: <290465.67505.qm@web37408.mail.mud.yahoo.com> Hi folks, yes, I'm doing it, stay tuned! :) We're in touch with Steve about this. Attila --- On Fri, 2/6/09, Steve Pieper wrote: > From: Steve Pieper > Subject: Re: [slicer-devel] tcl8.5 > To: "Stephen Aylward" > Cc: "slicer-devel at bwh.harvard.edu" > Date: Friday, February 6, 2009, 5:08 PM > Hi Stephen - > > The upgrade to tcl8.5 got stalled due to a problem building > BLT on the > mac. So I guess we'll stick with 8.4 for slicer3.4. > Attila is looking > at the cvs head of BLT (and new incrTcl/iwidgets) for his > solaris builds > so that might also work for the mac and we could upgrade > across the > board after the release branch. > > -Steve > > Stephen Aylward wrote: > > Hi, > > > > Tried enabling tcl8.5 via the option in getbuildtest, > and I get the > > following error: > > > > *** No rule to make target > > > `/home/aylward/src/DashboardSrcs/Slicer3/../Slicer3-lib/tcl-build/lib/libtcl8.4.so > > > ', needed by > `bin/libvtkCommonTCL.so.5.3.0'. Stop. > > > > > > I'll go back to using 8.4 > > > > s > > > > -- > > Stephen R. Aylward, Ph.D. > > Chief Medical Scientist > > Kitware, Inc. - North Carolina Office > > http://www.kitware.com > > (518) 371-3971 x300 > > > > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with > unsubscribe as the subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with > unsubscribe as the subject From pieper at bwh.harvard.edu Mon Feb 9 09:00:49 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 09 Feb 2009 09:00:49 -0500 Subject: [slicer-devel] Action Item: Documentation-3.4 In-Reply-To: <498C9362.90000@bwh.harvard.edu> References: <498C9362.90000@bwh.harvard.edu> Message-ID: <49903711.5030105@bwh.harvard.edu> Ron Kikinis wrote: > http://www.slicer.org/slicerWiki/index.php/Documentation-3.4 > ... > -The modules are organized hierarchically. Some of the modules are at > odd places for historic reasons. Please move your module where it belongs. > ... BTW - to put your module in a category you need to override the virtual GetCategory method for the module's GUI class, like this: http://viewvc.slicer.org/viewcvs.cgi/trunk/Modules/OpenIGTLinkIF/vtkOpenIGTLinkIFGUI.h?rev=8536&r1=8267&r2=8536 For scripted modules you can call the SetCategory method. -Steve From pieper at bwh.harvard.edu Tue Feb 10 12:20:57 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 10 Feb 2009 12:20:57 -0500 Subject: [slicer-devel] bug list review today 2pm 2nd floor 1249 Boylston Message-ID: <4991B779.4070601@bwh.harvard.edu> Let's get together today at 2 for a slicer developer meeting dedicated to reviewing the current outstanding bugs that we plan to fix for the 3.4 release. -Steve From pieper at bwh.harvard.edu Tue Feb 10 16:08:19 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 10 Feb 2009 16:08:19 -0500 Subject: [slicer-devel] Please review bug tracker In-Reply-To: <4991B779.4070601@bwh.harvard.edu> References: <4991B779.4070601@bwh.harvard.edu> Message-ID: <4991ECC3.1020004@bwh.harvard.edu> Hi - http://www.na-mic.org/Bug Developers: please review (and fix!) bugs that are assigned to you in the bug tracker. All: please check the status of bugs you have reported - many have been fixed but aren't closed until we get confirmation from the reporter. Please continue to report any bugs you find in nightly builds. Add notes to any existing bugs if you have new information (or just want to send a gentle reminder...) Thanks, Steve Steve Pieper wrote: > Let's get together today at 2 for a slicer developer meeting dedicated > to reviewing the current outstanding bugs that we plan to fix for the > 3.4 release. > > -Steve > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From kikinis at bwh.harvard.edu Wed Feb 11 07:53:18 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Wed, 11 Feb 2009 07:53:18 -0500 Subject: [slicer-devel] Action Item: Documentation for Slicer 3.4 In-Reply-To: <498C9362.90000@bwh.harvard.edu> References: <498C9362.90000@bwh.harvard.edu> Message-ID: <4992CA3E.50703@bwh.harvard.edu> Hi, I will now begin pushing for actually putting content in the pages. There are a number of people who have started with the work, the rest of you, please follow. I would like to see the documentation pages fully populated by Wednesday, 2/25. Please let me know, if you can not do that. Ron Ron Kikinis wrote: > http://www.slicer.org/slicerWiki/index.php/Documentation-3.4 > > Hello Slicer Developers, > > after the feature freeze for Slicer 3.4, I have generated a listing of > all Modules contained in the nightly build. We have added names for each > module as to who we think the owner of that module is. > > Please do the following by Tuesday 2/10: > -Review the list and see if your modules are listed. If you want them > removed, please do so. If I missed yours, and it should be listed, let > me know. > -The modules are organized hierarchically. Some of the modules are at > odd places for historic reasons. Please move your module where it belongs. > -Please send me an email to acknowledge that you have looked at this. > > And remember: If your module is listed, we will expect you to provide > documentation and other things as listed at the top of the page which is > linked above. > > Ron > -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From vrnova at gmail.com Fri Feb 13 11:36:27 2009 From: vrnova at gmail.com (Yanling Liu) Date: Fri, 13 Feb 2009 11:36:27 -0500 Subject: [slicer-devel] simple region growing module Message-ID: <24ceacd90902130836l88fd66cs32a003e0409c27a1@mail.gmail.com> Hello, may I get some help to locate source code for the simple region growing module in Slicer source tree? Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090213/b36547cb/attachment-0001.html From fedorov at bwh.harvard.edu Fri Feb 13 11:44:46 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Fri, 13 Feb 2009 11:44:46 -0500 Subject: [slicer-devel] simple region growing module In-Reply-To: <24ceacd90902130836l88fd66cs32a003e0409c27a1@mail.gmail.com> References: <24ceacd90902130836l88fd66cs32a003e0409c27a1@mail.gmail.com> Message-ID: <82f536810902130844v21391738ya3226ca44127caa8@mail.gmail.com> Applications/CLI/ConfidenceConnected.cxx On Fri, Feb 13, 2009 at 11:36 AM, Yanling Liu wrote: > Hello, may I get some help to locate source code for the simple region > growing module in Slicer source tree? > > Thanks, > Yanling > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > From curtislisle at knowledgevis.com Fri Feb 13 11:45:30 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Fri, 13 Feb 2009 11:45:30 -0500 Subject: [slicer-devel] simple region growing module In-Reply-To: <24ceacd90902130836l88fd66cs32a003e0409c27a1@mail.gmail.com> References: <24ceacd90902130836l88fd66cs32a003e0409c27a1@mail.gmail.com> Message-ID: <126CFB35-7246-42FC-850D-BF6C9B29579F@knowledgevis.com> Hi Yanling! it is hard to find, isn't it? The source is located in in Applications/CLI/ConfidenceConnected.cxx. The companion ConfidenceConnected.xml file that defines the GUI, provides a simpler name "Simple Region Growing" for the user. It is the XML file that provides the visible module names. Curt Curtis Lisle curtislisle at knowledgevis.com On Feb 13, 2009, at 11:36 AM, Yanling Liu wrote: > Hello, may I get some help to locate source code for the simple > region growing module in Slicer source tree? > > Thanks, > Yanling > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From kikinis at bwh.harvard.edu Fri Feb 13 14:10:09 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Fri, 13 Feb 2009 14:10:09 -0500 Subject: [slicer-devel] Documentation-3.4 Message-ID: <4995C591.1040800@bwh.harvard.edu> http://wiki.slicer.org/slicerWiki/index.php/Documentation-3.4#Modules Hi Slicer users and developers, we are in the process of preparing a new release of Slicer, Slicer 3.4. We have reorganized the modules and removed a number of modules. If you are a user, please let me know if we accidentally left out a module that you need and which was available in 3.2 or 3.3. If you are a developer, please let me know, if I removed your module and you want it back in the release. I would appreciate your feedback within a few days. Ron -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From vrnova at gmail.com Fri Feb 13 15:09:35 2009 From: vrnova at gmail.com (Yanling Liu) Date: Fri, 13 Feb 2009 15:09:35 -0500 Subject: [slicer-devel] quick question about modifying itk in Slicer-lib folder Message-ID: <24ceacd90902131209k415ae763gb7eab4f8cba08693@mail.gmail.com> Hi, if I modify itk source code in Slicer-lib folder and recompile itk, do I need to recompile Slicer to make my change take effect? I hope Slicer is dynamically liked to itk so I don't need to recompile Slicer every time. Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090213/81487f03/attachment.html From fedorov at bwh.harvard.edu Tue Feb 17 10:02:43 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 17 Feb 2009 10:02:43 -0500 Subject: [slicer-devel] Build problems with Visual Studio 9.0 Message-ID: <82f536810902170702q69d1a64bm4e91967451297ffc@mail.gmail.com> Hi, I have been experiencing problems building Slicer on Windows for some time now. I was wondering, if this is a known issue. 1). "PCbuild/pcbuild.sln /Updgrade" with the VS showing a message "Invalid Command Line. Unknown Switch : Upgrade." 2). after this I have error with python: running: {c:/Program Files/Microsoft Visual Studio 9.0/Common7/IDE/VCExpress.exe} PCbuild/pcbuild.sln /out buildlog.txt /build Release Slicer3_HOME is C:/cygwin/home/Andrey/Slicer3 child process exited abnormally I don't see this error compiling with Visual Studio 2003. Is this a known problem that I see? Is it possible to disable python build? I just need to compile Slicer on Windows so that I can use VC debugger. Thanks Andriy Fedorov From bill.lorensen at gmail.com Tue Feb 17 14:04:24 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Tue, 17 Feb 2009 14:04:24 -0500 Subject: [slicer-devel] Graffiti Alert: vtkFetchMIGUI.cxx warnings may be errors Message-ID: <4db4735c0902171104x4400751ds59fbee953ada23f5@mail.gmail.com> The warning: /home/lorensen/Projects/Slicer3Personal/Modules/FetchMI/vtkFetchMIGUI.cxx: In member function 'void vtkFetchMIGUI::AddModelNodes()': /home/lorensen/Projects/Slicer3Personal/Modules/FetchMI/vtkFetchMIGUI.cxx:693: warning: 'dtype' may be used uninitialized in this function applies to the code block: if ( fsnode != NULL ) { if ( snode->IsA("vtkMRMLFreeSurferModelStorageNode") ) { dtype = "FreeSurferModel"; } } else { dtype = "VTKModel"; } There is no else for the IsA test. What should dtype be is the IsA fails? /home/lorensen/Projects/Slicer3Personal/Modules/FetchMI/vtkFetchMIGUI.cxx: In member function 'void vtkFetchMIGUI::ShowSelectionTagView()': /home/lorensen/Projects/Slicer3Personal/Modules/FetchMI/vtkFetchMIGUI.cxx:1581: warning: 't' may be used uninitialized in this function For this one, looks like t should be initialized to NULL. Bill From pieper at bwh.harvard.edu Thu Feb 19 16:23:17 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 19 Feb 2009 16:23:17 -0500 Subject: [slicer-devel] Build problems with Visual Studio 9.0 In-Reply-To: <82f536810902170702q69d1a64bm4e91967451297ffc@mail.gmail.com> References: <82f536810902170702q69d1a64bm4e91967451297ffc@mail.gmail.com> Message-ID: <499DCDC5.7080605@bwh.harvard.edu> Hi Andriy - Yes, apparently the 'free' version of visual studio (VCExpress) doesn't support the /upgrade argument so we can't upgrade the python solution files on the fly. It looks like we'll need to find a workaround for this (like mirroring a set of upgraded solution files). For now you can turn USE_PYTHON off in slicer_variables.tcl. -Steve Andriy Fedorov wrote: > Hi, > > I have been experiencing problems building Slicer on Windows for some > time now. I was wondering, if this is a known issue. > > 1). "PCbuild/pcbuild.sln /Updgrade" with the VS showing a message > "Invalid Command Line. Unknown Switch : Upgrade." > > 2). after this I have error with python: > > running: {c:/Program Files/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpress.exe} PCbuild/pcbuild.sln /out buildlog.txt > /build Release > > Slicer3_HOME is C:/cygwin/home/Andrey/Slicer3 > child process exited abnormally > > I don't see this error compiling with Visual Studio 2003. > > Is this a known problem that I see? Is it possible to disable python > build? I just need to compile Slicer on Windows so that I can use VC > debugger. > > Thanks > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From fedorov at bwh.harvard.edu Fri Feb 20 09:17:04 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Fri, 20 Feb 2009 09:17:04 -0500 Subject: [slicer-devel] Build problems with Visual Studio 9.0 In-Reply-To: <499DCDC5.7080605@bwh.harvard.edu> References: <82f536810902170702q69d1a64bm4e91967451297ffc@mail.gmail.com> <499DCDC5.7080605@bwh.harvard.edu> Message-ID: <82f536810902200617w519e11f5m8a18fa3a614aa2e1@mail.gmail.com> Steve, This works. I think the problems I had before were because I was defining USE_PYTHON tp "off" in getbuildtest, and it was not propagated to genlib. The only thing I had to modify this time to get Slicer built was to comment out if { ![file exists bin/$::VTK_BUILD_TYPE/python25.dll] } { file copy $::Slicer3_LIB/python-build/PCbuild/python25.dll bin/$::VTK_BUILD_TYPE } in getbuildtest.tcl. Isn't USE_PYTHON check missing here? Thanks AF On Thu, Feb 19, 2009 at 4:23 PM, Steve Pieper wrote: > Hi Andriy - > > Yes, apparently the 'free' version of visual studio (VCExpress) doesn't > support the /upgrade argument so we can't upgrade the python solution files > on the fly. It looks like we'll need to find a workaround for this (like > mirroring a set of upgraded solution files). For now you can turn > USE_PYTHON off in slicer_variables.tcl. > > -Steve > > Andriy Fedorov wrote: >> >> Hi, >> >> I have been experiencing problems building Slicer on Windows for some >> time now. I was wondering, if this is a known issue. >> >> 1). "PCbuild/pcbuild.sln /Updgrade" with the VS showing a message >> "Invalid Command Line. Unknown Switch : Upgrade." >> >> 2). after this I have error with python: >> >> running: {c:/Program Files/Microsoft Visual Studio >> 9.0/Common7/IDE/VCExpress.exe} PCbuild/pcbuild.sln /out buildlog.txt >> /build Release >> >> Slicer3_HOME is C:/cygwin/home/Andrey/Slicer3 >> child process exited abnormally >> >> I don't see this error compiling with Visual Studio 2003. >> >> Is this a known problem that I see? Is it possible to disable python >> build? I just need to compile Slicer on Windows so that I can use VC >> debugger. >> >> Thanks >> >> Andriy Fedorov >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > From pieper at bwh.harvard.edu Fri Feb 20 09:44:31 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 20 Feb 2009 09:44:31 -0500 Subject: [slicer-devel] Build problems with Visual Studio 9.0 In-Reply-To: <82f536810902200617w519e11f5m8a18fa3a614aa2e1@mail.gmail.com> References: <82f536810902170702q69d1a64bm4e91967451297ffc@mail.gmail.com> <499DCDC5.7080605@bwh.harvard.edu> <82f536810902200617w519e11f5m8a18fa3a614aa2e1@mail.gmail.com> Message-ID: <499EC1CF.6090607@bwh.harvard.edu> Ah, yes - thanks I just wrapped that in an if statement and checked it in. -Steve Andriy Fedorov wrote: > Steve, > > This works. I think the problems I had before were because I was > defining USE_PYTHON tp "off" in getbuildtest, and it was not > propagated to genlib. > > The only thing I had to modify this time to get Slicer built was to comment out > > if { ![file exists bin/$::VTK_BUILD_TYPE/python25.dll] } { > file copy $::Slicer3_LIB/python-build/PCbuild/python25.dll > bin/$::VTK_BUILD_TYPE > } > > in getbuildtest.tcl. Isn't USE_PYTHON check missing here? > > Thanks > > AF > > > > > On Thu, Feb 19, 2009 at 4:23 PM, Steve Pieper wrote: >> Hi Andriy - >> >> Yes, apparently the 'free' version of visual studio (VCExpress) doesn't >> support the /upgrade argument so we can't upgrade the python solution files >> on the fly. It looks like we'll need to find a workaround for this (like >> mirroring a set of upgraded solution files). For now you can turn >> USE_PYTHON off in slicer_variables.tcl. >> >> -Steve >> >> Andriy Fedorov wrote: >>> Hi, >>> >>> I have been experiencing problems building Slicer on Windows for some >>> time now. I was wondering, if this is a known issue. >>> >>> 1). "PCbuild/pcbuild.sln /Updgrade" with the VS showing a message >>> "Invalid Command Line. Unknown Switch : Upgrade." >>> >>> 2). after this I have error with python: >>> >>> running: {c:/Program Files/Microsoft Visual Studio >>> 9.0/Common7/IDE/VCExpress.exe} PCbuild/pcbuild.sln /out buildlog.txt >>> /build Release >>> >>> Slicer3_HOME is C:/cygwin/home/Andrey/Slicer3 >>> child process exited abnormally >>> >>> I don't see this error compiling with Visual Studio 2003. >>> >>> Is this a known problem that I see? Is it possible to disable python >>> build? I just need to compile Slicer on Windows so that I can use VC >>> debugger. >>> >>> Thanks >>> >>> Andriy Fedorov >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject From kikinis at bwh.harvard.edu Sun Feb 22 12:01:12 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Sun, 22 Feb 2009 12:01:12 -0500 Subject: [slicer-devel] Action Item: Documentation for Slicer 3.4 In-Reply-To: <4992CA3E.50703@bwh.harvard.edu> References: <498C9362.90000@bwh.harvard.edu> <4992CA3E.50703@bwh.harvard.edu> Message-ID: <49A184D8.10903@bwh.harvard.edu> Just a reminder. Ron Ron Kikinis wrote: > Hi, > > I will now begin pushing for actually putting content in the pages. > There are a number of people who have started with the work, the rest of > you, please follow. > > I would like to see the documentation pages fully populated by > Wednesday, 2/25. Please let me know, if you can not do that. > > Ron > > Ron Kikinis wrote: >> http://www.slicer.org/slicerWiki/index.php/Documentation-3.4 >> >> Hello Slicer Developers, >> >> after the feature freeze for Slicer 3.4, I have generated a listing of >> all Modules contained in the nightly build. We have added names for >> each module as to who we think the owner of that module is. >> >> Please do the following by Tuesday 2/10: >> -Review the list and see if your modules are listed. If you want them >> removed, please do so. If I missed yours, and it should be listed, let >> me know. >> -The modules are organized hierarchically. Some of the modules are at >> odd places for historic reasons. Please move your module where it >> belongs. >> -Please send me an email to acknowledge that you have looked at this. >> >> And remember: If your module is listed, we will expect you to provide >> documentation and other things as listed at the top of the page which >> is linked above. >> >> Ron >> > -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From fedorov at bwh.harvard.edu Mon Feb 23 13:05:59 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 23 Feb 2009 13:05:59 -0500 Subject: [slicer-devel] On the use of SmartPointers Message-ID: <82f536810902231005u11aed444oa8cd4851a40e7113@mail.gmail.com> Hi, Tracking a memory leak in CommandLineModule, I discovered that it was caused by incorrect use of vtkSmartPointer, like this: vtkSmartPointer scene = vtkMRMLScene::New(); With this code, reference count becomes 2, and the scene never gets deallocated. Instead, it should be something like this vtkMRMLScene* scene = vtkMRMLScene::New(); vtkSmartPointer scenePtr = scene; scene->Delete(); Now, no additional Delete()'s on scenePtr are necessary. This example simply follows the rules of using SmartPointer's as explained on wiki here: http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Memory_Management#SmartPointers This fix eliminated most of memory leaks in CommandLineModule. Next, I searched through the code, and I found a number of other places where the same mistake is made. I send this to everybody as a warning -- if you are using SmartPointers in your code, you might want to check that you use them properly. Slicer experts, please correct me if I am mistaken... Andriy Fedorov From bill.lorensen at gmail.com Mon Feb 23 13:56:05 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Mon, 23 Feb 2009 13:56:05 -0500 Subject: [slicer-devel] On the use of SmartPointers In-Reply-To: <82f536810902231005u11aed444oa8cd4851a40e7113@mail.gmail.com> References: <82f536810902231005u11aed444oa8cd4851a40e7113@mail.gmail.com> Message-ID: <4db4735c0902231056p5397ee6dj15f048bbb744cd49@mail.gmail.com> Andriy, I think the proper pattern is: vtkSmartPointer scene = vtkSmartPointer::New(); If you do this, does the leak go away. The idea with smart pointers is that should should not have to delete them explicitly. Bill On Mon, Feb 23, 2009 at 1:05 PM, Andriy Fedorov wrote: > Hi, > > Tracking a memory leak in CommandLineModule, I discovered that it was > caused by incorrect use of vtkSmartPointer, like this: > > vtkSmartPointer scene = vtkMRMLScene::New(); > > With this code, reference count becomes 2, and the scene never gets > deallocated. Instead, it should be something like this > > vtkMRMLScene* scene = vtkMRMLScene::New(); > vtkSmartPointer scenePtr = scene; > scene->Delete(); > > Now, no additional Delete()'s on scenePtr are necessary. This example > simply follows the rules of using SmartPointer's as explained on wiki > here: http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Memory_Management#SmartPointers > > This fix eliminated most of memory leaks in CommandLineModule. Next, I > searched through the code, and I found a number of other places where > the same mistake is made. > > I send this to everybody as a warning -- if you are using > SmartPointers in your code, you might want to check that you use them > properly. > > Slicer experts, please correct me if I am mistaken... > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From fedorov at bwh.harvard.edu Mon Feb 23 14:24:23 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 23 Feb 2009 14:24:23 -0500 Subject: [slicer-devel] On the use of SmartPointers In-Reply-To: <4db4735c0902231056p5397ee6dj15f048bbb744cd49@mail.gmail.com> References: <82f536810902231005u11aed444oa8cd4851a40e7113@mail.gmail.com> <4db4735c0902231056p5397ee6dj15f048bbb744cd49@mail.gmail.com> Message-ID: <82f536810902231124m252a055cq128892c2b6868820@mail.gmail.com> Bill, I agree with you that SmartPointers should not be deleted explicitly, and the solution you provide is definitely more elegant. But the problem is that, at least as I understand, vtkSmartPointer scene = vtkSmartPointer::New(); is not the same as vtkSmartPointer scene = vtkMRMLScene::New(); In the snippet I showed, it's the scene that gets deleted, and after that SmartPointer to it is never deleted explicitly. I think if you do the second one, reference count will be increased once when the scene is created, and once again when the SmartPointer is assigned. If you look in Base/GUI/vtkSlicerSliceViewer.cxx:284, you will see this: vtkSmartPointer< vtkImageMapper > mapper = vtkImageMapper::New(); Is this the right use of SmartPointer? I believe it's not. But there is no memory leak, because we call mapper->Delete() in the next few lines! This indeed doesn't look like a proper use of SmartPointer. Fedorov On Mon, Feb 23, 2009 at 1:56 PM, Bill Lorensen wrote: > Andriy, > > I think the proper pattern is: > vtkSmartPointer scene = vtkSmartPointer::New(); > > If you do this, does the leak go away. The idea with smart pointers is > that should should not have to delete them explicitly. > > Bill > > On Mon, Feb 23, 2009 at 1:05 PM, Andriy Fedorov wrote: >> Hi, >> >> Tracking a memory leak in CommandLineModule, I discovered that it was >> caused by incorrect use of vtkSmartPointer, like this: >> >> vtkSmartPointer scene = vtkMRMLScene::New(); >> >> With this code, reference count becomes 2, and the scene never gets >> deallocated. Instead, it should be something like this >> >> vtkMRMLScene* scene = vtkMRMLScene::New(); >> vtkSmartPointer scenePtr = scene; >> scene->Delete(); >> >> Now, no additional Delete()'s on scenePtr are necessary. This example >> simply follows the rules of using SmartPointer's as explained on wiki >> here: http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Memory_Management#SmartPointers >> >> This fix eliminated most of memory leaks in CommandLineModule. Next, I >> searched through the code, and I found a number of other places where >> the same mistake is made. >> >> I send this to everybody as a warning -- if you are using >> SmartPointers in your code, you might want to check that you use them >> properly. >> >> Slicer experts, please correct me if I am mistaken... >> >> Andriy Fedorov >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > From bill.lorensen at gmail.com Mon Feb 23 14:43:47 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Mon, 23 Feb 2009 14:43:47 -0500 Subject: [slicer-devel] On the use of SmartPointers In-Reply-To: <82f536810902231124m252a055cq128892c2b6868820@mail.gmail.com> References: <82f536810902231005u11aed444oa8cd4851a40e7113@mail.gmail.com> <4db4735c0902231056p5397ee6dj15f048bbb744cd49@mail.gmail.com> <82f536810902231124m252a055cq128892c2b6868820@mail.gmail.com> Message-ID: <4db4735c0902231143s171bee49o9d29f773da84531b@mail.gmail.com> I think that SmartPointer is misused in many places in Slicer3. I don't think a Delete should ever be used if the SmartPoiinter is used as I suggested. I searched all of VTK and in all of the uses I say, they followed my pattern. Jim Miller is the best guy to answer this one. I may be way off. On Mon, Feb 23, 2009 at 2:24 PM, Andriy Fedorov wrote: > Bill, > > I agree with you that SmartPointers should not be deleted explicitly, > and the solution you provide is definitely more elegant. But the > problem is that, at least as I understand, > > vtkSmartPointer scene = vtkSmartPointer::New(); > > is not the same as > > vtkSmartPointer scene = vtkMRMLScene::New(); > > In the snippet I showed, it's the scene that gets deleted, and after > that SmartPointer to it is never deleted explicitly. > > I think if you do the second one, reference count will be increased > once when the scene is created, and once again when the SmartPointer > is assigned. > > If you look in Base/GUI/vtkSlicerSliceViewer.cxx:284, you will see this: > > vtkSmartPointer< vtkImageMapper > mapper = vtkImageMapper::New(); > > Is this the right use of SmartPointer? I believe it's not. But there > is no memory leak, because we call mapper->Delete() in the next few > lines! This indeed doesn't look like a proper use of SmartPointer. > > Fedorov > > > > > On Mon, Feb 23, 2009 at 1:56 PM, Bill Lorensen wrote: >> Andriy, >> >> I think the proper pattern is: >> vtkSmartPointer scene = vtkSmartPointer::New(); >> >> If you do this, does the leak go away. The idea with smart pointers is >> that should should not have to delete them explicitly. >> >> Bill >> >> On Mon, Feb 23, 2009 at 1:05 PM, Andriy Fedorov wrote: >>> Hi, >>> >>> Tracking a memory leak in CommandLineModule, I discovered that it was >>> caused by incorrect use of vtkSmartPointer, like this: >>> >>> vtkSmartPointer scene = vtkMRMLScene::New(); >>> >>> With this code, reference count becomes 2, and the scene never gets >>> deallocated. Instead, it should be something like this >>> >>> vtkMRMLScene* scene = vtkMRMLScene::New(); >>> vtkSmartPointer scenePtr = scene; >>> scene->Delete(); >>> >>> Now, no additional Delete()'s on scenePtr are necessary. This example >>> simply follows the rules of using SmartPointer's as explained on wiki >>> here: http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Memory_Management#SmartPointers >>> >>> This fix eliminated most of memory leaks in CommandLineModule. Next, I >>> searched through the code, and I found a number of other places where >>> the same mistake is made. >>> >>> I send this to everybody as a warning -- if you are using >>> SmartPointers in your code, you might want to check that you use them >>> properly. >>> >>> Slicer experts, please correct me if I am mistaken... >>> >>> Andriy Fedorov >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >> > From pieper at bwh.harvard.edu Mon Feb 23 15:18:23 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 23 Feb 2009 15:18:23 -0500 Subject: [slicer-devel] On the use of SmartPointers In-Reply-To: <4db4735c0902231143s171bee49o9d29f773da84531b@mail.gmail.com> References: <82f536810902231005u11aed444oa8cd4851a40e7113@mail.gmail.com> <4db4735c0902231056p5397ee6dj15f048bbb744cd49@mail.gmail.com> <82f536810902231124m252a055cq128892c2b6868820@mail.gmail.com> <4db4735c0902231143s171bee49o9d29f773da84531b@mail.gmail.com> Message-ID: <49A3048F.1090707@bwh.harvard.edu> Jim and I ran into this problem in the slice viewer last December and fixed the leaks but did not change the code to the corrected smart pointer style (Jim - thought /you/ were going to do that!). We did decide that this was the correct style: vtkSmartPointer< vtkImageMapper > mapper vtkSmartPointer::New(); -Steve Bill Lorensen wrote: > I think that SmartPointer is misused in many places in Slicer3. I > don't think a Delete should ever be used if the SmartPoiinter is used > as I suggested. I searched all of VTK and in all of the uses I say, > they followed my pattern. > > Jim Miller is the best guy to answer this one. I may be way off. > > On Mon, Feb 23, 2009 at 2:24 PM, Andriy Fedorov wrote: >> Bill, >> >> I agree with you that SmartPointers should not be deleted explicitly, >> and the solution you provide is definitely more elegant. But the >> problem is that, at least as I understand, >> >> vtkSmartPointer scene = vtkSmartPointer::New(); >> >> is not the same as >> >> vtkSmartPointer scene = vtkMRMLScene::New(); >> >> In the snippet I showed, it's the scene that gets deleted, and after >> that SmartPointer to it is never deleted explicitly. >> >> I think if you do the second one, reference count will be increased >> once when the scene is created, and once again when the SmartPointer >> is assigned. >> >> If you look in Base/GUI/vtkSlicerSliceViewer.cxx:284, you will see this: >> >> vtkSmartPointer< vtkImageMapper > mapper = vtkImageMapper::New(); >> >> Is this the right use of SmartPointer? I believe it's not. But there >> is no memory leak, because we call mapper->Delete() in the next few >> lines! This indeed doesn't look like a proper use of SmartPointer. >> >> Fedorov >> >> >> >> >> On Mon, Feb 23, 2009 at 1:56 PM, Bill Lorensen wrote: >>> Andriy, >>> >>> I think the proper pattern is: >>> vtkSmartPointer scene = vtkSmartPointer::New(); >>> >>> If you do this, does the leak go away. The idea with smart pointers is >>> that should should not have to delete them explicitly. >>> >>> Bill >>> >>> On Mon, Feb 23, 2009 at 1:05 PM, Andriy Fedorov wrote: >>>> Hi, >>>> >>>> Tracking a memory leak in CommandLineModule, I discovered that it was >>>> caused by incorrect use of vtkSmartPointer, like this: >>>> >>>> vtkSmartPointer scene = vtkMRMLScene::New(); >>>> >>>> With this code, reference count becomes 2, and the scene never gets >>>> deallocated. Instead, it should be something like this >>>> >>>> vtkMRMLScene* scene = vtkMRMLScene::New(); >>>> vtkSmartPointer scenePtr = scene; >>>> scene->Delete(); >>>> >>>> Now, no additional Delete()'s on scenePtr are necessary. This example >>>> simply follows the rules of using SmartPointer's as explained on wiki >>>> here: http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Memory_Management#SmartPointers >>>> >>>> This fix eliminated most of memory leaks in CommandLineModule. Next, I >>>> searched through the code, and I found a number of other places where >>>> the same mistake is made. >>>> >>>> I send this to everybody as a warning -- if you are using >>>> SmartPointers in your code, you might want to check that you use them >>>> properly. >>>> >>>> Slicer experts, please correct me if I am mistaken... >>>> >>>> Andriy Fedorov >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >>>> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From nagy.attila at yahoo.com Thu Feb 26 17:56:04 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Thu, 26 Feb 2009 14:56:04 -0800 (PST) Subject: [slicer-devel] File - Add Volume: problem parsing dicom dirs Message-ID: <875665.55405.qm@web37404.mail.mud.yahoo.com> Hi all, if you look at the pictures, you'll see what I mean. It happened between the 16 Feb and the 17th Feb. (I have a bunch of Solaris 32 bit builds in case anybody wants to try out - 28 of them, to be precise! :)) Same dataset, same machine. Is this Solaris-only, or something that remained unnoticed so far? Could somebody please take a look at it? Thanks, as always :) Attila -------------- next part -------------- A non-text attachment was scrubbed... Name: add.volume-ok.png Type: image/png Size: 71206 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090226/b71cd3f1/attachment-0001.png -------------- next part -------------- A non-text attachment was scrubbed... Name: bad.add.volume Type: application/octet-stream Size: 86936 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090226/b71cd3f1/attachment-0001.obj From INORTON at PARTNERS.ORG Thu Feb 26 19:05:28 2009 From: INORTON at PARTNERS.ORG (Norton, Isaiah) Date: Thu, 26 Feb 2009 19:05:28 -0500 Subject: [slicer-devel] File - Add Volume: problem parsing dicom dirs References: <875665.55405.qm@web37404.mail.mud.yahoo.com> Message-ID: <30A5D5C0F8C31B46AB1B8A7A3E18F93A0322BFA5@PHSXMB6.partners.org> Hi Attila, Steve fixed a few parser problems on 2/19. See this bug report: http://www.na-mic.org/Mantis/view.php?id=405 The ordering issue is fixed. There is still an issue with the parser (incorrectly) splitting out 1-3 files at a time from what should be a contiguous series -- that appears to be the problem for you, yes? I sent Steve a sample dataset and he is looking at it. -Isaiah -----Original Message----- From: slicer-devel-bounces at bwh.harvard.edu on behalf of Attila Nagy Sent: Thu 2/26/2009 5:56 PM To: Slicer Devel List Subject: [slicer-devel] File - Add Volume: problem parsing dicom dirs Hi all, if you look at the pictures, you'll see what I mean. It happened between the 16 Feb and the 17th Feb. (I have a bunch of Solaris 32 bit builds in case anybody wants to try out - 28 of them, to be precise! :)) Same dataset, same machine. Is this Solaris-only, or something that remained unnoticed so far? Could somebody please take a look at it? Thanks, as always :) Attila From nagy.attila at yahoo.com Thu Feb 26 19:40:13 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Thu, 26 Feb 2009 16:40:13 -0800 (PST) Subject: [slicer-devel] File - Add Volume: problem parsing dicom dirs In-Reply-To: <30A5D5C0F8C31B46AB1B8A7A3E18F93A0322BFA5@PHSXMB6.partners.org> Message-ID: <897614.19534.qm@web37403.mail.mud.yahoo.com> Hi Isaiah, Thanks for the answer! I happened to notice it today; in the morning I "upgraded" from 14-01-2009 to today's nightly, and there I saw this problem I began to step backwards day by day with the Slicer builds, and the problem disappeared when I went from 2/17's nightly to the 16th's nightly. I really don't know if this has something to do with changes from the 19th. I see the bug mentions something similar, but I don't know if it is really the same thing. Steve, if you need such dataset(s), just drop me a line; there are at least two in the range of 220MB-550MB :) Attila --- On Fri, 2/27/09, Norton, Isaiah wrote: > From: Norton, Isaiah > Subject: RE: [slicer-devel] File - Add Volume: problem parsing dicom dirs > To: "Attila Nagy" , "Slicer Devel List" > Date: Friday, February 27, 2009, 1:05 AM > Hi Attila, > Steve fixed a few parser problems on 2/19. See this bug > report: > > http://www.na-mic.org/Mantis/view.php?id=405 > > The ordering issue is fixed. There is still an issue with > the parser > (incorrectly) splitting out 1-3 files at a time from what > should be a contiguous > series -- that appears to be the problem for you, yes? I > sent Steve a sample > dataset and he is looking at it. > > -Isaiah > > > -----Original Message----- > From: slicer-devel-bounces at bwh.harvard.edu on behalf of > Attila Nagy > Sent: Thu 2/26/2009 5:56 PM > To: Slicer Devel List > Subject: [slicer-devel] File - Add Volume: problem parsing > dicom dirs > > Hi all, > > if you look at the pictures, you'll see what I mean. > It happened between the 16 Feb and the 17th Feb. (I have a > bunch of Solaris 32 > bit builds in case anybody wants to try out - 28 of them, > to be precise! :)) > Same dataset, same machine. > Is this Solaris-only, or something that remained unnoticed > so far? > > Could somebody please take a look at it? > > Thanks, as always :) > Attila > > > > > > > The information in this e-mail is intended only for the > person to whom it is > addressed. If you believe this e-mail was sent to you in > error and the e-mail > contains patient information, please contact the Partners > Compliance HelpLine at > http://www.partners.org/complianceline . If the e-mail was > sent to you in error > but does not contain patient information, please contact > the sender and properly > dispose of the e-mail. From pieper at bwh.harvard.edu Fri Feb 27 08:01:32 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 27 Feb 2009 08:01:32 -0500 Subject: [slicer-devel] File - Add Volume: problem parsing dicom dirs In-Reply-To: <897614.19534.qm@web37403.mail.mud.yahoo.com> References: <897614.19534.qm@web37403.mail.mud.yahoo.com> Message-ID: <49A7E42C.6080000@bwh.harvard.edu> Hi Attila, Isaiah - This is actually a feature ;) But maybe it's not so clear what's going on and we need a better design. In your example, the series "Thorax 5.0 B31f" is still available for loading as a single series with 128 files. But sub-series are also available for loading based on their unique ContentTime values. The rationale for this is that for some acquisitions, like time-series (e.g. MR dynamic contrast enhancement), there are multiple volumes in a series that are distinquised by ContentTime. Unfortunately, in CT scans it seems that a single volume is acquired at multiple ContentTimes. (There were a couple of bug reports around these topics, thus the big interface change). The subseries are currently divided based on these dicom fields: SeriesInstanceUID 0020|000E ContentTime 0008|0033 TriggerTime 0018|1060 DiffusionGradientOrientation 0018|9089 ImageOrientationPatient 0020|0037 So anytime the dialog displays something like " for " it is one of these 'virtual sub-series' with matching values that you can ignore or load as you wish. This new LoadVolume interface allows the user to choose between the full series or the sub-series. (The sub-series entries will always be lower in the tree). Some of us have been talking about how we might improve the display of this but nothing concrete yet... Since the mapping between a "dicom series" and a "slicer volume" is not one-to-one, perhaps we need to add more heuristics like 'ignore content time for CT'. Or add an options dialog for the parsing. Also, the new dialog can recursively search directories to find dicom files and sort them by patient/study/series/'sub-series'. There are still a few bugs with the new dialog and (Isaiah and Ron have given me some data to chew on). -Steve Attila Nagy wrote: > Hi Isaiah, > > Thanks for the answer! > I happened to notice it today; in the morning I "upgraded" from 14-01-2009 to today's nightly, and there I saw this problem > I began to step backwards day by day with the Slicer builds, and the problem disappeared when I went from 2/17's nightly to the 16th's nightly. > I really don't know if this has something to do with changes from the 19th. I see the bug mentions something similar, but I don't know if it is really the same thing. > Steve, if you need such dataset(s), just drop me a line; there are at least two in the range of 220MB-550MB :) > > > Attila > > --- On Fri, 2/27/09, Norton, Isaiah wrote: > >> From: Norton, Isaiah >> Subject: RE: [slicer-devel] File - Add Volume: problem parsing dicom dirs >> To: "Attila Nagy" , "Slicer Devel List" >> Date: Friday, February 27, 2009, 1:05 AM >> Hi Attila, >> Steve fixed a few parser problems on 2/19. See this bug >> report: >> >> http://www.na-mic.org/Mantis/view.php?id=405 >> >> The ordering issue is fixed. There is still an issue with >> the parser >> (incorrectly) splitting out 1-3 files at a time from what >> should be a contiguous >> series -- that appears to be the problem for you, yes? I >> sent Steve a sample >> dataset and he is looking at it. >> >> -Isaiah >> >> >> -----Original Message----- >> From: slicer-devel-bounces at bwh.harvard.edu on behalf of >> Attila Nagy >> Sent: Thu 2/26/2009 5:56 PM >> To: Slicer Devel List >> Subject: [slicer-devel] File - Add Volume: problem parsing >> dicom dirs >> >> Hi all, >> >> if you look at the pictures, you'll see what I mean. >> It happened between the 16 Feb and the 17th Feb. (I have a >> bunch of Solaris 32 >> bit builds in case anybody wants to try out - 28 of them, >> to be precise! :)) >> Same dataset, same machine. >> Is this Solaris-only, or something that remained unnoticed >> so far? >> >> Could somebody please take a look at it? >> >> Thanks, as always :) >> Attila >> >> >> >> >> >> >> The information in this e-mail is intended only for the >> person to whom it is >> addressed. If you believe this e-mail was sent to you in >> error and the e-mail >> contains patient information, please contact the Partners >> Compliance HelpLine at >> http://www.partners.org/complianceline . If the e-mail was >> sent to you in error >> but does not contain patient information, please contact >> the sender and properly >> dispose of the e-mail. > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From nagy.attila at yahoo.com Fri Feb 27 08:17:34 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Fri, 27 Feb 2009 05:17:34 -0800 (PST) Subject: [slicer-devel] File - Add Volume: problem parsing dicom dirs Message-ID: <577081.39853.qm@web37403.mail.mud.yahoo.com> Aha, it's clear now. I did notice that I could load the volumes as usual (sorry, I forgot to mention it earlier), and I thought that if a volume shows up twice, it's a thing to report :) Thanks, Attila --- On Fri, 2/27/09, Steve Pieper wrote: > From: Steve Pieper > Subject: Re: [slicer-devel] File - Add Volume: problem parsing dicom dirs > To: nagy.attila at yahoo.com > Cc: "Slicer Devel List" , "Norton, Isaiah" > Date: Friday, February 27, 2009, 2:01 PM > Hi Attila, Isaiah - > > This is actually a feature ;) But maybe it's not so > clear what's going on and we need a better design. > > In your example, the series "Thorax 5.0 B31f" is > still available for loading as a single series with 128 > files. But sub-series are also available for loading based > on their unique ContentTime values. > > The rationale for this is that for some acquisitions, like > time-series (e.g. MR dynamic contrast enhancement), there > are multiple volumes in a series that are distinquised by > ContentTime. Unfortunately, in CT scans it seems that a > single volume is acquired at multiple ContentTimes. (There > were a couple of bug reports around these topics, thus the > big interface change). > > The subseries are currently divided based on these dicom > fields: > > SeriesInstanceUID 0020|000E > ContentTime 0008|0033 > TriggerTime 0018|1060 > DiffusionGradientOrientation 0018|9089 > ImageOrientationPatient 0020|0037 > > So anytime the dialog displays something like > " for " it is > one of these 'virtual sub-series' with matching > values that you can ignore or load as you wish. This new > LoadVolume interface allows the user to choose between the > full series or the sub-series. (The sub-series entries will > always be lower in the tree). > > Some of us have been talking about how we might improve the > display of this but nothing concrete yet... Since the > mapping between a "dicom series" and a > "slicer volume" is not one-to-one, perhaps we need > to add more heuristics like 'ignore content time for > CT'. Or add an options dialog for the parsing. > > Also, the new dialog can recursively search directories to > find dicom files and sort them by > patient/study/series/'sub-series'. > > There are still a few bugs with the new dialog and (Isaiah > and Ron have given me some data to chew on). > > -Steve > > > > > Attila Nagy wrote: > > Hi Isaiah, > > > > Thanks for the answer! > > I happened to notice it today; in the morning I > "upgraded" from 14-01-2009 to today's nightly, > and there I saw this problem > > I began to step backwards day by day with the Slicer > builds, and the problem disappeared when I went from > 2/17's nightly to the 16th's nightly. > > I really don't know if this has something to do > with changes from the 19th. I see the bug mentions something > similar, but I don't know if it is really the same > thing. > > Steve, if you need such dataset(s), just drop me a > line; there are at least two in the range of 220MB-550MB :) > > > > > > Attila > > > > --- On Fri, 2/27/09, Norton, Isaiah > wrote: > > > >> From: Norton, Isaiah > >> Subject: RE: [slicer-devel] File - Add Volume: > problem parsing dicom dirs > >> To: "Attila Nagy" > , "Slicer Devel List" > > >> Date: Friday, February 27, 2009, 1:05 AM > >> Hi Attila, > >> Steve fixed a few parser problems on 2/19. See > this bug > >> report: > >> > >> http://www.na-mic.org/Mantis/view.php?id=405 > >> > >> The ordering issue is fixed. There is still an > issue with > >> the parser > >> (incorrectly) splitting out 1-3 files at a time > from what > >> should be a contiguous > >> series -- that appears to be the problem for you, > yes? I > >> sent Steve a sample > >> dataset and he is looking at it. > >> > >> -Isaiah > >> > >> > >> -----Original Message----- > >> From: slicer-devel-bounces at bwh.harvard.edu on > behalf of > >> Attila Nagy > >> Sent: Thu 2/26/2009 5:56 PM > >> To: Slicer Devel List > >> Subject: [slicer-devel] File - Add Volume: problem > parsing > >> dicom dirs > >> Hi all, > >> > >> if you look at the pictures, you'll see what I > mean. > >> It happened between the 16 Feb and the 17th Feb. > (I have a > >> bunch of Solaris 32 > >> bit builds in case anybody wants to try out - 28 > of them, > >> to be precise! :)) > >> Same dataset, same machine. > >> Is this Solaris-only, or something that remained > unnoticed > >> so far? > >> > >> Could somebody please take a look at it? > >> > >> Thanks, as always :) > >> Attila > >> > >> > >> > >> > >> > >> The information in this e-mail is intended only > for the > >> person to whom it is > >> addressed. If you believe this e-mail was sent to > you in > >> error and the e-mail > >> contains patient information, please contact the > Partners > >> Compliance HelpLine at > >> http://www.partners.org/complianceline . If the > e-mail was > >> sent to you in error > >> but does not contain patient information, please > contact > >> the sender and properly > >> dispose of the e-mail. > > > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with > unsubscribe as the subject From millerjv at gmail.com Sun Mar 1 21:42:45 2009 From: millerjv at gmail.com (millerjv) Date: Sun, 1 Mar 2009 21:42:45 -0500 Subject: [slicer-devel] Module order in menus Message-ID: <64F84042-0101-4D88-9FE8-4B9C21B34A9C@gmail.com> I added a facility to control the order that modules appear in the module menu. Currently, modules are listed alphabetically within each "category" or submenu. Modules that do not have a category specified will appear in the top section of the module menu. You can now specify an "index" that Slicer will use to order the modules within a given category. Modules are sorted within a category first by index, then by name. Two modules with the same index will appear in the list in alphabetical order. For built in and loadable modules, you can override the method GetIndex() to specify a sort position. For command line modules, you can use the tag. I put the index tag after the tag. Filtering.Denoising 3 I guess I could have done using an attribute instead Filtering.Denoising but I didn't think of that until I started writing this message. If people prefer an attribute over a new tag, I can change the code to use an attribute in place of a tag. Index is an unsigned short datatype, defaulting to the maximum value of an unsigned short. Therefore, if you do not specify an index for a module, it will appear in the menu after any module of the same category that has an index specified. Let me know if you have any issues. We can use this facility to get the modules listed in the module menu in the order that Ron requests. Jim From kikinis at bwh.harvard.edu Sun Mar 1 22:08:05 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Sun, 01 Mar 2009 22:08:05 -0500 Subject: [slicer-devel] Module order in menus In-Reply-To: <64F84042-0101-4D88-9FE8-4B9C21B34A9C@gmail.com> References: <64F84042-0101-4D88-9FE8-4B9C21B34A9C@gmail.com> Message-ID: <49AB4D95.5000903@bwh.harvard.edu> Thanks Jim. millerjv wrote: > I added a facility to control the order that modules appear in the > module menu. Currently, modules are listed alphabetically within each > "category" or submenu. Modules that do not have a category specified > will appear in the top section of the module menu. > > You can now specify an "index" that Slicer will use to order the > modules within a given category. Modules are sorted within a category > first by index, then by name. Two modules with the same index will > appear in the list in alphabetical order. > > For built in and loadable modules, you can override the method > GetIndex() to specify a sort position. > > For command line modules, you can use the tag. I put the > index tag after the tag. > > Filtering.Denoising > 3 > > I guess I could have done using an attribute instead > > Filtering.Denoising > > but I didn't think of that until I started writing this message. If > people prefer an attribute over a new tag, I can change the code to > use an attribute in place of a tag. > > Index is an unsigned short datatype, defaulting to the maximum value > of an unsigned short. Therefore, if you do not specify an index for a > module, it will appear in the menu after any module of the same > category that has an index specified. > > Let me know if you have any issues. We can use this facility to get > the modules listed in the module menu in the order that Ron requests. > > Jim > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From pieper at bwh.harvard.edu Mon Mar 2 08:13:44 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 02 Mar 2009 08:13:44 -0500 Subject: [slicer-devel] Module order in menus In-Reply-To: <49AB4D95.5000903@bwh.harvard.edu> References: <64F84042-0101-4D88-9FE8-4B9C21B34A9C@gmail.com> <49AB4D95.5000903@bwh.harvard.edu> Message-ID: <49ABDB88.1080606@bwh.harvard.edu> Hi Jim - looks great - small tweak: I added an instance variable for Index and a SetIndex method (so ScriptedModules can make use of this feature too). -Steve Ron Kikinis wrote: > Thanks Jim. > > millerjv wrote: >> I added a facility to control the order that modules appear in the >> module menu. Currently, modules are listed alphabetically within each >> "category" or submenu. Modules that do not have a category specified >> will appear in the top section of the module menu. >> >> You can now specify an "index" that Slicer will use to order the >> modules within a given category. Modules are sorted within a category >> first by index, then by name. Two modules with the same index will >> appear in the list in alphabetical order. >> >> For built in and loadable modules, you can override the method >> GetIndex() to specify a sort position. >> >> For command line modules, you can use the tag. I put the >> index tag after the tag. >> >> Filtering.Denoising >> 3 >> >> I guess I could have done using an attribute instead >> >> Filtering.Denoising >> >> but I didn't think of that until I started writing this message. If >> people prefer an attribute over a new tag, I can change the code to >> use an attribute in place of a tag. >> >> Index is an unsigned short datatype, defaulting to the maximum value >> of an unsigned short. Therefore, if you do not specify an index for a >> module, it will appear in the menu after any module of the same >> category that has an index specified. >> >> Let me know if you have any issues. We can use this facility to get >> the modules listed in the module menu in the order that Ron requests. >> >> Jim >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From nicole at bwh.harvard.edu Mon Mar 2 10:25:07 2009 From: nicole at bwh.harvard.edu (Nicole Aucoin) Date: Mon, 2 Mar 2009 10:25:07 -0500 (EST) Subject: [slicer-devel] Module order in menus In-Reply-To: <49ABDB88.1080606@bwh.harvard.edu> References: <64F84042-0101-4D88-9FE8-4B9C21B34A9C@gmail.com> <49AB4D95.5000903@bwh.harvard.edu> <49ABDB88.1080606@bwh.harvard.edu> Message-ID: For interactive modules in Slicer3/Modules, you currently add this line to your vtkModuleNameGUI.h file (inheriting from Base/GUI/vtkSlicerModuleGUI.h) to set the category: const char *GetCategory() const {return "MyCategory";} But it looks like you need to add this->Index = [myindex]; to your GUI constructor. Steve or Jim, could we make these match up so that the category and index are dealt with in the same way? I'd vote for making Category an instance variable (and possibly changing the variable names to CategoryName and CategoryIndex for clarity). Nicole On Mon, 2 Mar 2009, Steve Pieper wrote: > Hi Jim - looks great - small tweak: I added an instance variable for > Index and a SetIndex method (so ScriptedModules can make use of this > feature too). > > -Steve > > Ron Kikinis wrote: >> Thanks Jim. >> >> millerjv wrote: >>> I added a facility to control the order that modules appear in the >>> module menu. Currently, modules are listed alphabetically within each >>> "category" or submenu. Modules that do not have a category specified >>> will appear in the top section of the module menu. >>> >>> You can now specify an "index" that Slicer will use to order the >>> modules within a given category. Modules are sorted within a category >>> first by index, then by name. Two modules with the same index will >>> appear in the list in alphabetical order. >>> >>> For built in and loadable modules, you can override the method >>> GetIndex() to specify a sort position. >>> >>> For command line modules, you can use the tag. I put the >>> index tag after the tag. >>> >>> Filtering.Denoising >>> 3 >>> >>> I guess I could have done using an attribute instead >>> >>> Filtering.Denoising >>> >>> but I didn't think of that until I started writing this message. If >>> people prefer an attribute over a new tag, I can change the code to >>> use an attribute in place of a tag. >>> >>> Index is an unsigned short datatype, defaulting to the maximum value >>> of an unsigned short. Therefore, if you do not specify an index for a >>> module, it will appear in the menu after any module of the same >>> category that has an index specified. >>> >>> Let me know if you have any issues. We can use this facility to get >>> the modules listed in the module menu in the order that Ron requests. >>> >>> Jim >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > -- Nicole Aucoin nicole at bwh.harvard.edu Surgical Planning Laboratory, Brigham and Women's Hospital Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 http://www.slicer.org From kikinis at bwh.harvard.edu Mon Mar 2 10:39:42 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Mon, 02 Mar 2009 10:39:42 -0500 Subject: [slicer-devel] Module order in menus In-Reply-To: References: <64F84042-0101-4D88-9FE8-4B9C21B34A9C@gmail.com> <49AB4D95.5000903@bwh.harvard.edu> <49ABDB88.1080606@bwh.harvard.edu> Message-ID: <49ABFDBE.20000@bwh.harvard.edu> For Slicer 3.4 I would like to use the organization as laid out in the documentation webpage: http://www.slicer.org/slicerWiki/index.php/Documentation-3.4#Core_and_Loadable_Modules Nicole Aucoin wrote: > For interactive modules in Slicer3/Modules, you currently add this line > to your vtkModuleNameGUI.h file (inheriting from > Base/GUI/vtkSlicerModuleGUI.h) to set the category: > const char *GetCategory() const {return "MyCategory";} > But it looks like you need to add > this->Index = [myindex]; > to your GUI constructor. Steve or Jim, could we make these match up so > that the category and index are dealt with in the same way? I'd vote for > making Category an instance variable (and possibly changing the variable > names to CategoryName and CategoryIndex for clarity). > > Nicole > > On Mon, 2 Mar 2009, Steve Pieper wrote: > >> Hi Jim - looks great - small tweak: I added an instance variable for >> Index and a SetIndex method (so ScriptedModules can make use of this >> feature too). >> >> -Steve >> >> Ron Kikinis wrote: >>> Thanks Jim. >>> >>> millerjv wrote: >>>> I added a facility to control the order that modules appear in the >>>> module menu. Currently, modules are listed alphabetically within each >>>> "category" or submenu. Modules that do not have a category specified >>>> will appear in the top section of the module menu. >>>> >>>> You can now specify an "index" that Slicer will use to order the >>>> modules within a given category. Modules are sorted within a category >>>> first by index, then by name. Two modules with the same index will >>>> appear in the list in alphabetical order. >>>> >>>> For built in and loadable modules, you can override the method >>>> GetIndex() to specify a sort position. >>>> >>>> For command line modules, you can use the tag. I put the >>>> index tag after the tag. >>>> >>>> Filtering.Denoising >>>> 3 >>>> >>>> I guess I could have done using an attribute instead >>>> >>>> Filtering.Denoising >>>> >>>> but I didn't think of that until I started writing this message. If >>>> people prefer an attribute over a new tag, I can change the code to >>>> use an attribute in place of a tag. >>>> >>>> Index is an unsigned short datatype, defaulting to the maximum value >>>> of an unsigned short. Therefore, if you do not specify an index for a >>>> module, it will appear in the menu after any module of the same >>>> category that has an index specified. >>>> >>>> Let me know if you have any issues. We can use this facility to get >>>> the modules listed in the module menu in the order that Ron requests. >>>> >>>> Jim >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From curtislisle at knowledgevis.com Mon Mar 2 10:59:24 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Mon, 2 Mar 2009 10:59:24 -0500 Subject: [slicer-devel] CLI argument passing - where to look in the code? Message-ID: <412DE3C9-1498-4C4C-B34C-6981D0BFCED4@knowledgevis.com> Where is the code that describes how to get Slicer internal information passed on the command line? The specific problem I want to learn about is if Slicer's new region of interest boundaries can be passed on the command line to future CLI modules. Thanks for any suggestions. Curt From fedorov at bwh.harvard.edu Mon Mar 2 11:14:41 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 2 Mar 2009 11:14:41 -0500 Subject: [slicer-devel] CLI argument passing - where to look in the code? In-Reply-To: <412DE3C9-1498-4C4C-B34C-6981D0BFCED4@knowledgevis.com> References: <412DE3C9-1498-4C4C-B34C-6981D0BFCED4@knowledgevis.com> Message-ID: <82f536810903020814o235c9c9bkf34334e040211ebd@mail.gmail.com> Curt, Possibly function ApplyTask() in Modules/CommandLineModule/vtkCommandLineModuleLogic.cxx is what you are looking for. It's a long long long function, which figures out all possible combinations of inputs and different ways CLI can be implemented and executed. If you look around line 826 in the code, it has a piece about regions, but I cannot tell you anything more than this, because I never used this particular functionality. Hope this helps Fedorov On Mon, Mar 2, 2009 at 10:59 AM, Curtis Lisle wrote: > Where is the code that describes how to get Slicer internal > information passed on the command line? ?The specific problem I want > to learn about is if Slicer's new region of interest boundaries can be > passed on the command line to future CLI modules. ?Thanks for any > suggestions. > > Curt > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From pieper at bwh.harvard.edu Mon Mar 2 11:16:58 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 02 Mar 2009 11:16:58 -0500 Subject: [slicer-devel] Module order in menus In-Reply-To: References: <64F84042-0101-4D88-9FE8-4B9C21B34A9C@gmail.com> <49AB4D95.5000903@bwh.harvard.edu> <49ABDB88.1080606@bwh.harvard.edu> Message-ID: <49AC067A.2090309@bwh.harvard.edu> Hi Nicole - I had made Category an instance variable at the vtkScripteModuleGUI level, but you're right, it makes more sense at the vtkSlicerModuleGUI level for consistency with the Index. I'll make that change and check it in. -Steve Nicole Aucoin wrote: > For interactive modules in Slicer3/Modules, you currently add this line > to your vtkModuleNameGUI.h file (inheriting from > Base/GUI/vtkSlicerModuleGUI.h) to set the category: > const char *GetCategory() const {return "MyCategory";} > But it looks like you need to add > this->Index = [myindex]; > to your GUI constructor. Steve or Jim, could we make these match up so > that the category and index are dealt with in the same way? I'd vote for > making Category an instance variable (and possibly changing the variable > names to CategoryName and CategoryIndex for clarity). > > Nicole > > On Mon, 2 Mar 2009, Steve Pieper wrote: > >> Hi Jim - looks great - small tweak: I added an instance variable for >> Index and a SetIndex method (so ScriptedModules can make use of this >> feature too). >> >> -Steve >> >> Ron Kikinis wrote: >>> Thanks Jim. >>> >>> millerjv wrote: >>>> I added a facility to control the order that modules appear in the >>>> module menu. Currently, modules are listed alphabetically within each >>>> "category" or submenu. Modules that do not have a category specified >>>> will appear in the top section of the module menu. >>>> >>>> You can now specify an "index" that Slicer will use to order the >>>> modules within a given category. Modules are sorted within a category >>>> first by index, then by name. Two modules with the same index will >>>> appear in the list in alphabetical order. >>>> >>>> For built in and loadable modules, you can override the method >>>> GetIndex() to specify a sort position. >>>> >>>> For command line modules, you can use the tag. I put the >>>> index tag after the tag. >>>> >>>> Filtering.Denoising >>>> 3 >>>> >>>> I guess I could have done using an attribute instead >>>> >>>> Filtering.Denoising >>>> >>>> but I didn't think of that until I started writing this message. If >>>> people prefer an attribute over a new tag, I can change the code to >>>> use an attribute in place of a tag. >>>> >>>> Index is an unsigned short datatype, defaulting to the maximum value >>>> of an unsigned short. Therefore, if you do not specify an index for a >>>> module, it will appear in the menu after any module of the same >>>> category that has an index specified. >>>> >>>> Let me know if you have any issues. We can use this facility to get >>>> the modules listed in the module menu in the order that Ron requests. >>>> >>>> Jim >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > From vrnova at gmail.com Mon Mar 2 15:00:47 2009 From: vrnova at gmail.com (Yanling Liu) Date: Mon, 2 Mar 2009 15:00:47 -0500 Subject: [slicer-devel] label map extraction Message-ID: <24ceacd90903021200v4645d5feo98f10e7e30e99f79@mail.gmail.com> Hello, is there a module in Slicer that will allow me to extract different label maps from one volume and save them into different new volumes? For example, support I have a volume including 10 different label maps and now I want to create 10 new volumes so that the first label map goes into the first new volume, the second label map goes into the second new volume, etc... Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090302/e86cbfd9/attachment.html From fedorov at bwh.harvard.edu Mon Mar 2 15:04:29 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 2 Mar 2009 15:04:29 -0500 Subject: [slicer-devel] label map extraction In-Reply-To: <24ceacd90903021200v4645d5feo98f10e7e30e99f79@mail.gmail.com> References: <24ceacd90903021200v4645d5feo98f10e7e30e99f79@mail.gmail.com> Message-ID: <82f536810903021204g4565fe1fh9745dc5c4c9e8217@mail.gmail.com> Yanling, You can use Threshold Image module (from the Filtering category) to extract label(s), and then save them as new images. AF On Mon, Mar 2, 2009 at 3:00 PM, Yanling Liu wrote: > Hello, is there a module in Slicer that will allow me to extract different > label maps from one volume and save them into different new volumes? > > For example, support I have a volume including 10 different label maps and > now I want to create 10 new volumes so that the first label map goes into > the first new volume, the second label map goes into the second new volume, > etc... > > Thanks, > Yanling > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > From vrnova at gmail.com Tue Mar 3 10:51:26 2009 From: vrnova at gmail.com (Yanling Liu) Date: Tue, 3 Mar 2009 10:51:26 -0500 Subject: [slicer-devel] ERROR: input scans can not be aligned Message-ID: <24ceacd90903030751rbd36a42u788c6bdd9e55b051@mail.gmail.com> Hello, when testing the ChangeTracker module with two input volumes I get an error message says "input scans can not be aligned". The first volume is the original scan with spherical lung tumor. In the second volume I applied Gaussian filter to grow the tumor a little bit. These two volumes has same parameters such as dimension, origin, position and spacings so the error message confuses me.... Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090303/60e7c659/attachment.html From vrnova at gmail.com Tue Mar 3 11:55:45 2009 From: vrnova at gmail.com (Yanling Liu) Date: Tue, 3 Mar 2009 11:55:45 -0500 Subject: [slicer-devel] queryatlas Message-ID: <24ceacd90903030855s79b32bf3w92d7aaf08a21d489@mail.gmail.com> Hello, is there a way to create interactive annotation for regular polygon vtk models (.vtk files) using QueryAtlas module? I can see the tool tip string of the "Set up annotations:" button says currently supports FreeSurfer models only... Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090303/6c9e5fe3/attachment.html From wjp at bwh.harvard.edu Tue Mar 3 12:19:22 2009 From: wjp at bwh.harvard.edu (Wendy Plesniak) Date: Tue, 03 Mar 2009 12:19:22 -0500 Subject: [slicer-devel] queryatlas In-Reply-To: <24ceacd90903030855s79b32bf3w92d7aaf08a21d489@mail.gmail.com> References: <24ceacd90903030855s79b32bf3w92d7aaf08a21d489@mail.gmail.com> Message-ID: <49AD669A.8070602@bwh.harvard.edu> Hi Yanling, currently the queryatlas doesn't annotate vtk polygon models; only freesurfer datasets with an aparc.annot file, and label maps with semantic label names. The functionality you mention is on our to-do list, however we probably won't get to it for some time. cheers, -wen Yanling Liu wrote: > Hello, is there a way to create interactive annotation for regular > polygon vtk models (.vtk files) using QueryAtlas module? I can see the > tool tip string of the "Set up annotations:" button says currently > supports FreeSurfer models only... > > Thanks, > Yanling > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From fedorov at bwh.harvard.edu Tue Mar 3 13:30:24 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 3 Mar 2009 13:30:24 -0500 Subject: [slicer-devel] ERROR: input scans can not be aligned In-Reply-To: <24ceacd90903030751rbd36a42u788c6bdd9e55b051@mail.gmail.com> References: <24ceacd90903030751rbd36a42u788c6bdd9e55b051@mail.gmail.com> Message-ID: <82f536810903031030y3ffea398oa4890199d6337346@mail.gmail.com> Yanling, The error probably was "ERROR: Failed to align input scans!", not "can not be aligned". This error is a generic error that means rigid registration failed for one or another reason. This is explained in the ChangeTracker documentation: http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 The parameters for registration in ChangeTracker were tuned for the meningioma brain MRI, and if registration fails, there is no mechanism to re-adjust the parameters. I understand it should not fail, because images are the same, but that is what I believe is happening. Can you try to run RigidRegistration module with the default parameters on that data you have and see if it completes without errors? Please let me know what happens if you try what I suggest. Andriy Fedorov On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: > Hello, when testing the ChangeTracker module with two input volumes I get an > error message says "input scans can not be aligned". The first volume is the > original scan with spherical lung tumor. In the second volume I applied > Gaussian filter to grow the tumor a little bit. These two volumes? has same > parameters such as dimension, origin, position and spacings so the error > message confuses me.... > > Thanks, > Yanling > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > From simon at math.uh.edu Tue Mar 3 15:39:57 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Tue, 3 Mar 2009 14:39:57 -0600 Subject: [slicer-devel] re-orienting planes Message-ID: <200903031439.57712.simon@math.uh.edu> Hi, I'm wondering if someone can help me understand this behavior; otherwise I expect I'll have to grovel the various layers of slicer code involved for ages to sort it out. I have some TCL code to re-orient slices, and I've noticed some strange behavior. I've isolated it here. The code I ran this time uses the SetSliceToRASByNTP in vtkMRMLSliceNode.cxx, but I see the basically same thing when I compute all the matrices myself. Here's what's happening. Say I start the axial/sagital/coronal slice orientations and start rotating the coronal slice, I'll paste here the state of the transform matrices at each step as I introduce the new orientation. It mostly looks like I expected except the UpdateMatrices call results in the position part of the SliceToRAS matrix being different than what was passed to SetSliceToRASByNTP. I've attached some debug output as "aligned.txt" However, if my starting point is a reformat type slice in an arbitrary orientation, I can get something like the attached "arbitrary.txt". In this case, not only has the position moved, the normal has changed and so has the vector "t". Something in UpdateMatrices is rotating and translating the slice from the location given to SetSliceToRASByNTP. Now this method seems to work with the gui widget for reorienting the planes, so I suspect I just haven't understood something happening in a lower layer. Perhaps it is a more fundamental mistake, though. This is proving a bit frustrating to pin down, so I'd really appreciate the insight of anyone familiar with the code base. I first though I was just misunderstanding some convention, but the matrices all seem to check out ok. I can trigger the same behavior by doing things myself (rather than SetSliceToRASByNTP) and just setting Modified() on the slice afterward.... cheers, Simon -------------- next part -------------- ###### planes axis aligned ######## *** ORIGINAL *** SliceToRAS -1 -0 0 0.550774 -0 -0 1 1.00797 0 1 0 0.445251 -0 0 -0 1 XYToSlice 0.57661 0 0 -122.818 0 0.576611 0 -102.925 0 0 1.00797 0 0 0 0 1 XYToRAS -0.57661 0 0 123.369 0 0 1.00797 1.00797 0 0.576611 0 -102.48 0 0 0 1 norm t = 1.0 norm c = 1.0 norm n = 1.0 t dot c = 0.0 c dot n = 0.0 n dot t = 0.0 SliceToRAS*XYToSlice - XYToRAS = 0.0 nx: -0.0998335 ny: 0.995004 nz: 0.0 tx: -0.995004 ty: -0.0998335 tz: 0.0 px: 0.26438 py: 2.85731 pz: 0.445251 *** SetSliceToRASByNTP *** SliceToRAS -0.995004 0 -0.0998335 0.279451 -0.0998335 0 0.995004 2.70711 0 1 0 0.445251 -0 0 -0 1 XYToSlice 0.57661 0 0 -122.818 0 0.576611 0 -102.925 0 0 1.00797 0 0 0 0 1 XYToRAS -0.57373 0 -0.100629 122.484 -0.057565 0 1.00293 14.9685 0 0.576611 0 -102.48 0 0 0 1 norm t = 1.0 norm c = 1.0 norm n = 1.0 t dot c = 0.0 c dot n = 0.0 n dot t = -3.1834e-08 SliceToRAS*XYToSlice - XYToRAS = 0.0 -------------- next part -------------- ###### planes arbitrarily aligned ######## *** ORIGINAL *** SliceToRAS -0.827236 0.0735319 -0.557029 -0.238702 -0.55315 -0.28052 0.784437 33.9435 -0.0985765 0.957034 0.272727 3.03299 -0 0 -0 1 XYToSlice 0.57661 0 0 -122.818 0 0.576611 0 -102.925 0 0 0.892963 0 0 0 0 1 XYToRAS -0.476993 0.0423993 -0.497406 93.7925 -0.318952 -0.161751 0.700473 130.753 -0.0568402 0.551836 0.243535 -83.3628 0 0 0 1 norm t = 1.0 norm c = 1.0 norm n = 1.0 t dot c = 3.37899e-07 c dot n = -9.88282e-07 n dot t = -1.77312e-06 SliceToRAS*XYToSlice - XYToRAS = 0.0 nx: -0.634089 ny: 0.721099 nz: 0.279196 tx: -0.768639 ty: -0.627215 tz: -0.125726 px: -0.982944 py: 34.8842 pz: 3.37782 *** SetSliceToRASByNTP *** SliceToRAS -0.768635 0.0844544 -0.634088 1.17658 -0.627211 -0.29432 0.721098 32.4283 -0.125725 0.951968 0.279196 2.42696 -0 0 -0 1 XYToSlice 0.57661 0 0 -122.818 0 0.576611 0 -102.925 0 0 0.892963 0 0 0 0 1 XYToRAS -0.443203 0.0486973 -0.566217 86.8863 -0.361656 -0.169708 0.643914 139.754 -0.0724944 0.548915 0.249311 -80.113 0 0 0 1 norm t = 1.0 norm c = 1.0 norm n = 1.0 t dot c = -1.768e-07 c dot n = 6.57728e-07 n dot t = -5.171e-07 SliceToRAS*XYToSlice - XYToRAS = 0.0 From fedorov at bwh.harvard.edu Wed Mar 4 09:16:03 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Wed, 4 Mar 2009 09:16:03 -0500 Subject: [slicer-devel] Changes in crosshair GUI Message-ID: <82f536810903040616s5ff9a507ue38a6d4e7943981f@mail.gmail.com> Hi, Starting this Sunday, I noticed that crosshair setup has changed in Slicer. At startup, there is no crosshair, and if I select basic crosshair from the GUI (or any other type), crosshair does not follow the mouse, but stays in the center of slice views. Today's nightly build has the same issue. Is this a bug, or a new feature? Andriy Fedorov From karthik.krishnan at kitware.com Wed Mar 4 16:16:45 2009 From: karthik.krishnan at kitware.com (Karthik Krishnan) Date: Wed, 4 Mar 2009 16:16:45 -0500 Subject: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] build issues Message-ID: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> Hello: Is the following module available for use/testing in the Slicer repository, NAMIC Sandbox or elsewhere ? http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation I get a build errors building the TRProstateBiopsy module in the NAMICSandbox : In file included from /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.h:28:35: error: vtkSlicerVolumesLogic.h: No such file or directory Not all directories in Slicer are provided as include directories for this module, which makes me think that there must be some other recommended way of building this module. I just tried building NAMICSandBox/IGTLoadableModules/TRProstateBiopsy, setting Slicer3_DIR to my build of Slicer3. Any help would be appreciated. Thanks Regards -- Karthik Krishnan R&D Engineer, Kitware Inc. Ph: 518 371 3971 x119 Fax: 518 371 3971 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090304/2cd00201/attachment.html From pieper at bwh.harvard.edu Wed Mar 4 17:52:22 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 04 Mar 2009 17:52:22 -0500 Subject: [slicer-devel] re-orienting planes In-Reply-To: <200903031439.57712.simon@math.uh.edu> References: <200903031439.57712.simon@math.uh.edu> Message-ID: <49AF0626.5030708@bwh.harvard.edu> Hi Simon - I think I see what's going on but I'm having a little trouble following the steps. Do you think you could provide a short tcl script that reproduces the issue? (perhaps an excerpt of what you are using?). It should look something like this: set sliceNode [$::slicer3::MRMLScene GetNthNodeByClass 0 "vtkMRMLSliceNode"] $sliceNode GetSliceToRAS [$sliceNode GetSliceToRAS] Identity $sliceNode SetSliceToRASByNTP $sliceNode UpdateMatrices ... (further manipulate sliceNode...) Thanks, Steve Simon K Alexander wrote: > Hi, > > I'm wondering if someone can help me understand this behavior; otherwise I expect I'll have to grovel the various layers > of slicer code involved for ages to sort it out. > > I have some TCL code to re-orient slices, and I've noticed some strange behavior. I've isolated it here. The code I > ran this time uses the SetSliceToRASByNTP in vtkMRMLSliceNode.cxx, but I see the basically same thing when I compute > all the matrices myself. > > Here's what's happening. Say I start the axial/sagital/coronal slice orientations and start rotating the coronal slice, > I'll paste here the state of the transform matrices at each step as I introduce the new orientation. It mostly looks > like I expected except the UpdateMatrices call results in the position part of the SliceToRAS matrix being different > than what was passed to SetSliceToRASByNTP. > > I've attached some debug output as "aligned.txt" > > However, if my starting point is a reformat type slice in an arbitrary orientation, I can get something like the > attached "arbitrary.txt". In this case, not only has the position moved, the normal has changed and so has the > vector "t". Something in UpdateMatrices is rotating and translating the slice from the location given to > SetSliceToRASByNTP. > > Now this method seems to work with the gui widget for reorienting the planes, so I suspect I just haven't understood > something happening in a lower layer. Perhaps it is a more fundamental mistake, though. > > This is proving a bit frustrating to pin down, so I'd really appreciate the insight of anyone familiar with the code > base. I first though I was just misunderstanding some convention, but the matrices all seem to check out ok. I can > trigger the same behavior by doing things myself (rather than SetSliceToRASByNTP) and just setting Modified() on the > slice afterward.... > > > cheers, > Simon > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Wed Mar 4 19:17:47 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 04 Mar 2009 19:17:47 -0500 Subject: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] build issues In-Reply-To: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> References: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> Message-ID: <49AF1A2B.9030209@bwh.harvard.edu> Hi Karthik - I don't know much about that module, but from the build error looks like you may need to add the volumes module to your include_directories command (these moved a while ago but maybe this module was never updated). include_directories( .... ${Volumes_SOURCE_DIR} ${Volumes_BINARY_DIR} ) Good luck, Steve Karthik Krishnan wrote: > Hello: > > Is the following module available for use/testing in the Slicer > repository, NAMIC Sandbox or elsewhere ? > > http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation > > > I get a build errors building the TRProstateBiopsy module in the > NAMICSandbox : > > In file included from > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.h:28:35: > error: vtkSlicerVolumesLogic.h: No such file or directory > > Not all directories in Slicer are provided as include directories for > this module, which makes me think that there must be some other > recommended way of building this module. > > I just tried building NAMICSandBox/IGTLoadableModules/TRProstateBiopsy, > setting Slicer3_DIR to my build of Slicer3. > > Any help would be appreciated. > > Thanks > Regards > > -- > Karthik Krishnan > R&D Engineer, > Kitware Inc. > Ph: 518 371 3971 x119 > Fax: 518 371 3971 > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From karthik.krishnan at kitware.com Wed Mar 4 21:06:25 2009 From: karthik.krishnan at kitware.com (Karthik Krishnan) Date: Wed, 4 Mar 2009 21:06:25 -0500 Subject: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] build issues In-Reply-To: <49AF1A2B.9030209@bwh.harvard.edu> References: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> <49AF1A2B.9030209@bwh.harvard.edu> Message-ID: <9ddb27260903041806v224df2c9ua5057254f3ab1f2d@mail.gmail.com> Thanks Steve. That's what I've been doing :) Adding include directories. Still get a host of build errors. I'll post more results after poking around soon. Thanks Regards -- karthik On Wed, Mar 4, 2009 at 7:17 PM, Steve Pieper wrote: > Hi Karthik - > > I don't know much about that module, but from the build error looks like > you may need to add the volumes module to your include_directories command > (these moved a while ago but maybe this module was never updated). > > include_directories( > .... > ${Volumes_SOURCE_DIR} > ${Volumes_BINARY_DIR} > ) > > Good luck, > Steve > > Karthik Krishnan wrote: > >> Hello: >> >> Is the following module available for use/testing in the Slicer >> repository, NAMIC Sandbox or elsewhere ? >> >> http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation >> >> >> I get a build errors building the TRProstateBiopsy module in the >> NAMICSandbox : >> >> In file included from >> /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: >> /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.h:28:35: >> error: vtkSlicerVolumesLogic.h: No such file or directory >> >> Not all directories in Slicer are provided as include directories for this >> module, which makes me think that there must be some other recommended way >> of building this module. >> >> I just tried building NAMICSandBox/IGTLoadableModules/TRProstateBiopsy, >> setting Slicer3_DIR to my build of Slicer3. >> >> Any help would be appreciated. >> >> Thanks >> Regards >> >> -- >> Karthik Krishnan >> R&D Engineer, >> Kitware Inc. >> Ph: 518 371 3971 x119 >> Fax: 518 371 3971 >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >> > -- Karthik Krishnan R&D Engineer, Kitware Inc. Ph: 518 371 3971 x119 Fax: 518 371 3971 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090304/8b4b6852/attachment.html From fedorov at bwh.harvard.edu Thu Mar 5 09:23:57 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Thu, 5 Mar 2009 09:23:57 -0500 Subject: [slicer-devel] Rigid registration of aligned images Message-ID: <82f536810903050623q7908ef76u4ca02d853bff96e4@mail.gmail.com> Hi all, This is in relation to the previous post from Yanling http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html I attach the images Yanling sent to me, which create problems during rigid registration. Note, the second image is simply the first image after Gaussian blur. Both images have the same size, origin, orientation and spacing. If I use the default registration parameters, it fails with "ERROR: MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF summed to zero" I tried to choose the parameters to adapt to the input data: I reduced the number of samples to the minimum possible (1000), since the images are relatively small (81x77x11), reduced the translation scaling and number of iterations. Here are the parameters: Histogram bins: 30 Spatial samples: 1000 Iterations: 100,100,50,20 Learning rates: default settings Translation scaling: 10 Now registration fails with the message "ERROR: MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many samples map outside moving image buffer: 231 / 1000" How is it possible, if the images seemingly have 100% overlap? I tried different combinations of parameters, but none seems to work for these images. Can anyone explain what is going on? Jim, can you help? Thank you Andriy Fedorov On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov wrote: > Yanling, > > The error probably was "ERROR: Failed to align input scans!", not "can > not be aligned". > > This error is a generic error that means rigid registration failed for > one or another reason. This is explained in the ChangeTracker > documentation: > > http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 > > The parameters for registration in ChangeTracker were tuned for the > meningioma brain MRI, and if registration fails, there is no mechanism > to re-adjust the parameters. I understand it should not fail, because > images are the same, but that is what I believe is happening. > > Can you try to run RigidRegistration module with the default > parameters on that data you have and see if it completes without > errors? > > Please let me know what happens if you try what I suggest. > > Andriy Fedorov > > > > On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: >> Hello, when testing the ChangeTracker module with two input volumes I get an >> error message says "input scans can not be aligned". The first volume is the >> original scan with spherical lung tumor. In the second volume I applied >> Gaussian filter to grow the tumor a little bit. These two volumes? has same >> parameters such as dimension, origin, position and spacings so the error >> message confuses me.... >> >> Thanks, >> Yanling >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> > -------------- next part -------------- A non-text attachment was scrubbed... Name: croppedLung.nrrd Type: application/octet-stream Size: 81934 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090305/a7015c7d/attachment-0002.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: GaussianBlurCroppedLung.nrrd Type: application/octet-stream Size: 245551 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090305/a7015c7d/attachment-0003.obj From karthik.krishnan at kitware.com Thu Mar 5 10:36:00 2009 From: karthik.krishnan at kitware.com (Karthik Krishnan) Date: Thu, 5 Mar 2009 10:36:00 -0500 Subject: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] build issues In-Reply-To: <000301c99da5$8a5f35f0$9f1da1d0$@queensu.ca> References: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> <49AF1A2B.9030209@bwh.harvard.edu> <9ddb27260903041806v224df2c9ua5057254f3ab1f2d@mail.gmail.com> <1ab4d3450903050600u23f13e79nd367b695dcc02576@mail.gmail.com> <000301c99da5$8a5f35f0$9f1da1d0$@queensu.ca> Message-ID: <9ddb27260903050736l42742e9k2e5c5bc5556f3eb8@mail.gmail.com> Siddharth: Thanks for the quick response. In trying to build the module from its new location, it appears that it will build only on windows. I get build errors on my linux machine such as : In file included from /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.cxx:1: /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.h:7:21: error: Windows.h: No such file or directory Upon closer examination of the code, it doesn't appear that there is anything windows specific in the workflow/functionality itself. However, the code has been written to includet he vtkWin32OpenGLRenderWindow, instead of using the vtkRenderWindow and taking advantage of VTK's abstraction. Could you please tell us if the module windows specific ? Thanks -- karthik On Thu, Mar 5, 2009 at 10:17 AM, Siddharth Vikal wrote: > Hi Karthik, > > After reading the emails below, I did svn diff of CMakeLists.txt (working > base vs working copy). And all the include directories are there! > > Then, I looked closer in the email exchange, and it comes to my attention > that you are looking at the wrong location for the module. > > The correct location is here: > > *http://svn.na-mic.org/NAMICSandBox/trunk/Queens/TRProstateBiopsy/* > > instead of, as you quote. > > >>> In file included from > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > >>> > > I believe the above location is very very very old, which was created by > David Gobbi over an year ago. The module was moved to /Queens/TRProstateBiopsy > in June 2008 project week, and we have been committing to that location. > > I would request you not to waste further time to correct errors, but to > access the module from the above new location. I have added this location > the project page on wiki: > > *http://www.na-mic.org/Wiki/index.php/DBP2:JHU:Roadmap* > > I?ll commit the latest version right now. > > Let me know if there are queries or if run into problems this time. > > Best regards > > Siddharth > - Show quoted text - > > -----Original Message----- > From: gaoyi.cn at gmail.com [mailto:gaoyi.cn at gmail.com ] > On Behalf Of Yi Gao > Sent: Thursday, March 05, 2009 9:00 AM > To: Karthik Krishnan > Cc: Steve Pieper; slicer-devel at bwh.harvard.edu; Siddharth Vikal > Subject: Re: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] > build issues > > Hi Karthik, > > I'm sorry that I knew little about this module.... but I think > > Siddharth (CCed) may help on that. > > Best, > > yi > > On Wed, Mar 4, 2009 at 9:06 PM, Karthik Krishnan > > wrote: > > > Thanks Steve. That's what I've been doing :) Adding include directories. > > > Still get a host of build errors. I'll post more results after poking > around > > > soon. > > > > > > Thanks > > > Regards > > > -- > > > karthik > > > - Show quoted text - > > > > > > > > > On Wed, Mar 4, 2009 at 7:17 PM, Steve Pieper > wrote: > > >> > > >> Hi Karthik - > > >> > > >> I don't know much about that module, but from the build error looks like > > >> you may need to add the volumes module to your include_directories > command > > >> (these moved a while ago but maybe this module was never updated). > > >> > > >> include_directories( > > >> .... > > >> ${Volumes_SOURCE_DIR} > > >> ${Volumes_BINARY_DIR} > > >> ) > > >> > > >> Good luck, > > >> Steve > > >> > > >> Karthik Krishnan wrote: > > >>> > > >>> Hello: > > >>> > > >>> Is the following module available for use/testing in the Slicer > > >>> repository, NAMIC Sandbox or elsewhere ? > > >>> > > >>> http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation > > >>> > > >>> > > >>> I get a build errors building the TRProstateBiopsy module in the > > >>> NAMICSandbox : > > >>> > > >>> In file included from > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > >>> > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.h:28:35: > > >>> error: vtkSlicerVolumesLogic.h: No such file or directory > > >>> > > >>> Not all directories in Slicer are provided as include directories for > > >>> this module, which makes me think that there must be some other > recommended > > >>> way of building this module. > > >>> > > >>> I just tried building NAMICSandBox/IGTLoadableModules/TRProstateBiopsy, > > >>> setting Slicer3_DIR to my build of Slicer3. > > >>> > > >>> Any help would be appreciated. > > >>> > > >>> Thanks > > >>> Regards > > >>> > > >>> -- > > >>> Karthik Krishnan > > >>> R&D Engineer, > > >>> Kitware Inc. > > >>> Ph: 518 371 3971 x119 > > >>> Fax: 518 371 3971 > > >>> > > >>> > > >>> > ------------------------------------------------------------------------ > > >>> > > >>> _______________________________________________ > > >>> slicer-devel mailing list > > >>> slicer-devel at bwh.harvard.edu > > >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > >>> To unsubscribe: send email to > > >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > > >>> subject > > > > > > > > > > > > -- > > > Karthik Krishnan > > > R&D Engineer, > > > Kitware Inc. > > > Ph: 518 371 3971 x119 > > > Fax: 518 371 3971 > > > > -- Karthik Krishnan R&D Engineer, Kitware Inc. Ph: 518 371 3971 x119 Fax: 518 371 3971 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090305/c066d4a7/attachment.html From karthik.krishnan at kitware.com Thu Mar 5 11:01:01 2009 From: karthik.krishnan at kitware.com (Karthik Krishnan) Date: Thu, 5 Mar 2009 11:01:01 -0500 Subject: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] build issues In-Reply-To: <001801c99dab$3c69e0b0$b53da210$@queensu.ca> References: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> <49AF1A2B.9030209@bwh.harvard.edu> <9ddb27260903041806v224df2c9ua5057254f3ab1f2d@mail.gmail.com> <1ab4d3450903050600u23f13e79nd367b695dcc02576@mail.gmail.com> <000301c99da5$8a5f35f0$9f1da1d0$@queensu.ca> <9ddb27260903050736l42742e9k2e5c5bc5556f3eb8@mail.gmail.com> <001801c99dab$3c69e0b0$b53da210$@queensu.ca> Message-ID: <9ddb27260903050801i3ae3adb0g2470865d6989c7db@mail.gmail.com> Perfect. I'll wait for your heads up. Thanks -- karthik On Thu, Mar 5, 2009 at 10:58 AM, Siddharth Vikal wrote: > Hi Karthik, > > > > You are right. As of now, the module is actually windows specific. The > module?s objective is integrate with the MRI-compatible robotic device > developed at Johns Hopkins. That robot has sensors, and we have only windows > specific drivers for them. You can find that specific code in this file: > vtkTRProstateBiopsyUSBOpticalEncoder.h, .cxx > > > > I?m going to commit latest code in 15-20 minutes, and I?ll give you a heads > up. > > > > Thanks > > Siddharth > > > > *From:* Karthik Krishnan [mailto:karthik.krishnan at kitware.com] > *Sent:* Thursday, March 05, 2009 10:36 AM > *To:* Siddharth Vikal > *Cc:* Steve Pieper; slicer-devel at bwh.harvard.edu; Yi Gao- Show quoted text > - > > *Subject:* Re: [slicer-devel] Prostate segmentation [TRProstateBiopsy > module] build issues > - Show quoted text - > > > > Siddharth: > > Thanks for the quick response. > > In trying to build the module from its new location, it appears that it > will build only on windows. I get build errors on my linux machine such as : > > In file included from > /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.cxx:1: > /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.h:7:21: > error: Windows.h: No such file or directory > > > Upon closer examination of the code, it doesn't appear that there is > anything windows specific in the workflow/functionality itself. However, the > code has been written to includet he vtkWin32OpenGLRenderWindow, instead of > using the vtkRenderWindow and taking advantage of VTK's abstraction. > > Could you please tell us if the module windows specific ? > > Thanks > -- > karthik > > On Thu, Mar 5, 2009 at 10:17 AM, Siddharth Vikal > wrote: > > Hi Karthik, > > After reading the emails below, I did svn diff of CMakeLists.txt (working > base vs working copy). And all the include directories are there! > > Then, I looked closer in the email exchange, and it comes to my attention > that you are looking at the wrong location for the module. > > The correct location is here: > > http://svn.na-mic.org/NAMICSandBox/trunk/Queens/TRProstateBiopsy/ > > instead of, as you quote. > > >>> In file included from > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > >>> > > I believe the above location is very very very old, which was created by > David Gobbi over an year ago. The module was moved to > /Queens/TRProstateBiopsy in June 2008 project week, and we have been > committing to that location. > > I would request you not to waste further time to correct errors, but to > access the module from the above new location. I have added this location > the project page on wiki: > > http://www.na-mic.org/Wiki/index.php/DBP2:JHU:Roadmap > > I?ll commit the latest version right now. > > Let me know if there are queries or if run into problems this time. > > Best regards > > Siddharth > > - Show quoted text - > > -----Original Message----- > From: gaoyi.cn at gmail.com [mailto:gaoyi.cn at gmail.com ] > On Behalf Of Yi Gao > Sent: Thursday, March 05, 2009 9:00 AM > To: Karthik Krishnan > Cc: Steve Pieper; slicer-devel at bwh.harvard.edu; Siddharth Vikal > Subject: Re: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] > build issues > > Hi Karthik, > > I'm sorry that I knew little about this module.... but I think > > Siddharth (CCed) may help on that. > > Best, > > yi > > On Wed, Mar 4, 2009 at 9:06 PM, Karthik Krishnan > > wrote: > > > Thanks Steve. That's what I've been doing :) Adding include directories. > > > Still get a host of build errors. I'll post more results after poking > around > > > soon. > > > > > > Thanks > > > Regards > > > -- > > > karthik > > > - Show quoted text - > > > > > > > > > On Wed, Mar 4, 2009 at 7:17 PM, Steve Pieper > wrote: > > >> > > >> Hi Karthik - > > >> > > >> I don't know much about that module, but from the build error looks like > > >> you may need to add the volumes module to your include_directories > command > > >> (these moved a while ago but maybe this module was never updated). > > >> > > >> include_directories( > > >> .... > > >> ${Volumes_SOURCE_DIR} > > >> ${Volumes_BINARY_DIR} > > >> ) > > >> > > >> Good luck, > > >> Steve > > >> > > >> Karthik Krishnan wrote: > > >>> > > >>> Hello: > > >>> > > >>> Is the following module available for use/testing in the Slicer > > >>> repository, NAMIC Sandbox or elsewhere ? > > >>> > > >>> http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation > > >>> > > >>> > > >>> I get a build errors building the TRProstateBiopsy module in the > > >>> NAMICSandbox : > > >>> > > >>> In file included from > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > >>> > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.h:28:35: > > >>> error: vtkSlicerVolumesLogic.h: No such file or directory > > >>> > > >>> Not all directories in Slicer are provided as include directories for > > >>> this module, which makes me think that there must be some other > recommended > > >>> way of building this module. > > >>> > > >>> I just tried building NAMICSandBox/IGTLoadableModules/TRProstateBiopsy, > > >>> setting Slicer3_DIR to my build of Slicer3. > > >>> > > >>> Any help would be appreciated. > > >>> > > >>> Thanks > > >>> Regards > > >>> > > >>> -- > > >>> Karthik Krishnan > > >>> R&D Engineer, > > >>> Kitware Inc. > > >>> Ph: 518 371 3971 x119 > > >>> Fax: 518 371 3971 > > >>> > > >>> > > >>> > ------------------------------------------------------------------------ > > >>> > > >>> _______________________________________________ > > >>> slicer-devel mailing list > > >>> slicer-devel at bwh.harvard.edu > > >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > >>> To unsubscribe: send email to > > >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > > >>> subject > > > > > > > > > > > > -- > > > Karthik Krishnan > > > R&D Engineer, > > > Kitware Inc. > > > Ph: 518 371 3971 x119 > > > Fax: 518 371 3971 > > > > > > > > -- > Karthik Krishnan > R&D Engineer, > Kitware Inc. > Ph: 518 371 3971 x119 > Fax: 518 371 3971 > -- Karthik Krishnan R&D Engineer, Kitware Inc. Ph: 518 371 3971 x119 Fax: 518 371 3971 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090305/ece20c80/attachment-0001.html From simon at math.uh.edu Thu Mar 5 11:52:22 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Thu, 5 Mar 2009 10:52:22 -0600 Subject: [slicer-devel] re-orienting planes In-Reply-To: <49AF0626.5030708@bwh.harvard.edu> References: <200903031439.57712.simon@math.uh.edu> <49AF0626.5030708@bwh.harvard.edu> Message-ID: <200903051052.23329.simon@math.uh.edu> Hi Steve, Well, I've narrowed it down quite a bit. There were two things going on. As I noted in the previous reply, the example I sent to the list did not exhibit the rotation of the plane I had mentioned (bad bug report!), only translations of the point p. So I sorted out that this rotation was my fault, it only happened in a code path that was erroneously failing to keep XYToSlice strictly diagonal in the upper left 3x3 matrix, so when slicer code corrected this somewhere, it rotated the plane. In many cases this was nearly correct anyway, so the effect was subtle. However, this has nothing to do with the main problem I was having which is that the plane would "jump" sometimes. It turns out this is another instance of bug 0000410 I had reported in Mantis. The spurious rotations due to my error were throwing me off, so I didn't see it at first. I've attached a complete test case in "debug.zip" If you load the .mrml file, fire up the TCL interactor and "source debug.tcl" it will set everything up. You'll see that the last line of debug.tcl is rotations::rotate-slicenumber 0 -0.1 $p0 $p1 If you repeat this line in the interactor, you'll see the plane rotate around the indicated axis. However, note as you do it that the slice planes slider is always moving to the right in its range. After five repetitions (six including the one called in debug.tcl) the slider will be on the far right of the range. The next repetition causes the plane to translate to fit into its newly constrained range. You'll note at this point moving the slider across its entire range only moves the slice a small fraction of the total volume. So it looks like I've got the bug pinned down. Do you have a good idea where the code causing bug 0000410 is located? Can I help sort this out if you point me at the right place? This bug is really holding me back at the moment, but I feel like it might be a lot of code to read to locate it. I'll try anyway as time allows, but any assistance is much appreciated. cheers, Simon On Wednesday 04 March 2009 04:52:22 pm Steve Pieper wrote: > Hi Simon - > > I think I see what's going on but I'm having a little trouble following > the steps. Do you think you could provide a short tcl script that > reproduces the issue? (perhaps an excerpt of what you are using?). > > It should look something like this: > > set sliceNode [$::slicer3::MRMLScene GetNthNodeByClass 0 > "vtkMRMLSliceNode"] $sliceNode GetSliceToRAS > [$sliceNode GetSliceToRAS] Identity > $sliceNode SetSliceToRASByNTP > $sliceNode UpdateMatrices > ... (further manipulate sliceNode...) > > Thanks, > Steve > > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > > subject -------------- next part -------------- A non-text attachment was scrubbed... Name: debug.zip Type: application/x-zip Size: 3120683 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090305/178e7528/attachment-0001.bin From karthik.krishnan at kitware.com Thu Mar 5 12:42:55 2009 From: karthik.krishnan at kitware.com (Karthik Krishnan) Date: Thu, 5 Mar 2009 12:42:55 -0500 Subject: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] build issues In-Reply-To: <002c01c99db0$fa119130$ee34b390$@queensu.ca> References: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> <49AF1A2B.9030209@bwh.harvard.edu> <9ddb27260903041806v224df2c9ua5057254f3ab1f2d@mail.gmail.com> <1ab4d3450903050600u23f13e79nd367b695dcc02576@mail.gmail.com> <000301c99da5$8a5f35f0$9f1da1d0$@queensu.ca> <9ddb27260903050736l42742e9k2e5c5bc5556f3eb8@mail.gmail.com> <001801c99dab$3c69e0b0$b53da210$@queensu.ca> <9ddb27260903050801i3ae3adb0g2470865d6989c7db@mail.gmail.com> <002c01c99db0$fa119130$ee34b390$@queensu.ca> Message-ID: <9ddb27260903050942x418faec2mb982b6b980d21980@mail.gmail.com> Siddharth: Thanks for your help. I just committed a few minor fixes to get it to compile on linux. They are listed below. Please review them when you get chance. TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx - rev 3850. TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx - rev 3850. TRProstateBiopsy/CMakeLists.txt - rev 3850. Now that we are past the compilation, I have a second barrage of questions on using the module :) My sole interest is in the prostate segmentation part of the module. It appears that I need a "Seed Volume". What does the seed volume contain. Is it a binary UCHAR image with a couple of ON voxels within the prostate ? It would be far more intuitive to use fiducials on the prostate as illustrated in http://wiki.na-mic.org/Wiki/images/e/ee/KarthikMRPhantomWithEndorectalCoilBalloon.jpg --- Secondly, what is the role of the "Beta" parameter in the segmentation module. It would help if there was more documentation. Perhaps a dataset with a step by step walkthrough that one could try. The only documentation that I can find on the segmentation algorithm comes from the RandomWalker page here: http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation Thanks again for your help Regards -- karthik On Thu, Mar 5, 2009 at 11:39 AM, Siddharth Vikal wrote: > I?ve just committed the latest module version. > > > > But the module is still windows specific in some places for the reasons > mentioned. > > > > Thanks > > Siddharth > > > > *From:* Karthik Krishnan [mailto:karthik.krishnan at kitware.com] > *Sent:* Thursday, March 05, 2009 11:01 AM > *To:* Siddharth Vikal > *Cc:* Steve Pieper; slicer-devel at bwh.harvard.edu; Yi Gao > - Show quoted text - > *Subject:* Re: [slicer-devel] Prostate segmentation [TRProstateBiopsy > module] build issues > - Show quoted text - > > > > Perfect. I'll wait for your heads up. > > Thanks > -- > karthik > > On Thu, Mar 5, 2009 at 10:58 AM, Siddharth Vikal > wrote: > > Hi Karthik, > > > > You are right. As of now, the module is actually windows specific. The > module?s objective is integrate with the MRI-compatible robotic device > developed at Johns Hopkins. That robot has sensors, and we have only windows > specific drivers for them. You can find that specific code in this file: > vtkTRProstateBiopsyUSBOpticalEncoder.h, .cxx > > > > I?m going to commit latest code in 15-20 minutes, and I?ll give you a heads > up. > > > > Thanks > > Siddharth > > > > *From:* Karthik Krishnan [mailto:karthik.krishnan at kitware.com] > *Sent:* Thursday, March 05, 2009 10:36 AM > *To:* Siddharth Vikal > *Cc:* Steve Pieper; slicer-devel at bwh.harvard.edu; Yi Gao > > - Show quoted text - > > > *Subject:* Re: [slicer-devel] Prostate segmentation [TRProstateBiopsy > module] build issues > > - Show quoted text - > > > > Siddharth: > > Thanks for the quick response. > > In trying to build the module from its new location, it appears that it > will build only on windows. I get build errors on my linux machine such as : > > In file included from > /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.cxx:1: > /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.h:7:21: > error: Windows.h: No such file or directory > > > Upon closer examination of the code, it doesn't appear that there is > anything windows specific in the workflow/functionality itself. However, the > code has been written to includet he vtkWin32OpenGLRenderWindow, instead of > using the vtkRenderWindow and taking advantage of VTK's abstraction. > > Could you please tell us if the module windows specific ? > > Thanks > -- > karthik > > On Thu, Mar 5, 2009 at 10:17 AM, Siddharth Vikal > wrote: > > Hi Karthik, > > After reading the emails below, I did svn diff of CMakeLists.txt (working > base vs working copy). And all the include directories are there! > > Then, I looked closer in the email exchange, and it comes to my attention > that you are looking at the wrong location for the module. > > The correct location is here: > > http://svn.na-mic.org/NAMICSandBox/trunk/Queens/TRProstateBiopsy/ > > instead of, as you quote. > > >>> In file included from > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > >>> > > I believe the above location is very very very old, which was created by > David Gobbi over an year ago. The module was moved to > /Queens/TRProstateBiopsy in June 2008 project week, and we have been > committing to that location. > > I would request you not to waste further time to correct errors, but to > access the module from the above new location. I have added this location > the project page on wiki: > > http://www.na-mic.org/Wiki/index.php/DBP2:JHU:Roadmap > > I?ll commit the latest version right now. > > Let me know if there are queries or if run into problems this time. > > Best regards > > Siddharth > > - Show quoted text - > > -----Original Message----- > From: gaoyi.cn at gmail.com [mailto:gaoyi.cn at gmail.com ] > On Behalf Of Yi Gao > Sent: Thursday, March 05, 2009 9:00 AM > To: Karthik Krishnan > Cc: Steve Pieper; slicer-devel at bwh.harvard.edu; Siddharth Vikal > Subject: Re: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] > build issues > > Hi Karthik, > > I'm sorry that I knew little about this module.... but I think > > Siddharth (CCed) may help on that. > > Best, > > yi > > On Wed, Mar 4, 2009 at 9:06 PM, Karthik Krishnan > > wrote: > > > Thanks Steve. That's what I've been doing :) Adding include directories. > > > Still get a host of build errors. I'll post more results after poking > around > > > soon. > > > > > > Thanks > > > Regards > > > -- > > > karthik > > > - Show quoted text - > > > > > > > > > On Wed, Mar 4, 2009 at 7:17 PM, Steve Pieper > wrote: > > >> > > >> Hi Karthik - > > >> > > >> I don't know much about that module, but from the build error looks like > > >> you may need to add the volumes module to your include_directories > command > > >> (these moved a while ago but maybe this module was never updated). > > >> > > >> include_directories( > > >> .... > > >> ${Volumes_SOURCE_DIR} > > >> ${Volumes_BINARY_DIR} > > >> ) > > >> > > >> Good luck, > > >> Steve > > >> > > >> Karthik Krishnan wrote: > > >>> > > >>> Hello: > > >>> > > >>> Is the following module available for use/testing in the Slicer > > >>> repository, NAMIC Sandbox or elsewhere ? > > >>> > > >>> http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation > > >>> > > >>> > > >>> I get a build errors building the TRProstateBiopsy module in the > > >>> NAMICSandbox : > > >>> > > >>> In file included from > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > >>> > > >>> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.h:28:35: > > >>> error: vtkSlicerVolumesLogic.h: No such file or directory > > >>> > > >>> Not all directories in Slicer are provided as include directories for > > >>> this module, which makes me think that there must be some other > recommended > > >>> way of building this module. > > >>> > > >>> I just tried building NAMICSandBox/IGTLoadableModules/TRProstateBiopsy, > > >>> setting Slicer3_DIR to my build of Slicer3. > > >>> > > >>> Any help would be appreciated. > > >>> > > >>> Thanks > > >>> Regards > > >>> > > >>> -- > > >>> Karthik Krishnan > > >>> R&D Engineer, > > >>> Kitware Inc. > > >>> Ph: 518 371 3971 x119 > > >>> Fax: 518 371 3971 > > >>> > > >>> > > >>> > ------------------------------------------------------------------------ > > >>> > > >>> _______________________________________________ > > >>> slicer-devel mailing list > > >>> slicer-devel at bwh.harvard.edu > > >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > >>> To unsubscribe: send email to > > >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > > >>> subject > > > > > > > > > > > > -- > > > Karthik Krishnan > > > R&D Engineer, > > > Kitware Inc. > > > Ph: 518 371 3971 x119 > > > Fax: 518 371 3971 > > > > > > > > -- > Karthik Krishnan > R&D Engineer, > Kitware Inc. > Ph: 518 371 3971 x119 > Fax: 518 371 3971 > > > > > -- > Karthik Krishnan > R&D Engineer, > Kitware Inc. > Ph: 518 371 3971 x119 > Fax: 518 371 3971 > -- Karthik Krishnan R&D Engineer, Kitware Inc. Ph: 518 371 3971 x119 Fax: 518 371 3971 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090305/abf0d588/attachment-0001.html From kikinis at bwh.harvard.edu Thu Mar 5 12:54:46 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Thu, 05 Mar 2009 12:54:46 -0500 Subject: [slicer-devel] Prostate segmentation [TRProstateBiopsy module] build issues In-Reply-To: <9ddb27260903050942x418faec2mb982b6b980d21980@mail.gmail.com> References: <9ddb27260903041316p61af0d6egc393d60fd7cf3d95@mail.gmail.com> <49AF1A2B.9030209@bwh.harvard.edu> <9ddb27260903041806v224df2c9ua5057254f3ab1f2d@mail.gmail.com> <1ab4d3450903050600u23f13e79nd367b695dcc02576@mail.gmail.com> <000301c99da5$8a5f35f0$9f1da1d0$@queensu.ca> <9ddb27260903050736l42742e9k2e5c5bc5556f3eb8@mail.gmail.com> <001801c99dab$3c69e0b0$b53da210$@queensu.ca> <9ddb27260903050801i3ae3adb0g2470865d6989c7db@mail.gmail.com> <002c01c99db0$fa119130$ee34b390$@queensu.ca> <9ddb27260903050942x418faec2mb982b6b980d21980@mail.gmail.com> Message-ID: <49B011E6.1090306@bwh.harvard.edu> Hi, if this is part of the prostate nav module, then http://wiki.slicer.org/slicerWiki/index.php/Modules:ProstateNav-Documentation-3.4 is where the reference-manual-style documentation should go. And yes, please use the fiducials if it makes sense to use them. There is a lot of infrastructure supporting them. Check out the fiducials module documentation ;) http://wiki.slicer.org/slicerWiki/index.php/Modules:Fiducials-Documentation-3.4 Tutorials are currently in http://www.na-mic.org/Wiki/index.php/Slicer3.2:Training You could use 1.6 as a template Ron Karthik Krishnan wrote: > Siddharth: > > Thanks for your help. > > I just committed a few minor fixes to get it to compile on linux. They > are listed below. Please review them when you get chance. > > TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx - rev 3850. > TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx - rev 3850. > TRProstateBiopsy/CMakeLists.txt - rev 3850. > > Now that we are past the compilation, I have a second barrage of > questions on using the module :) My sole interest is in the prostate > segmentation part of the module. > > It appears that I need a "Seed Volume". What does the seed volume > contain. Is it a binary UCHAR image with a couple of ON voxels within > the prostate ? > > It would be far more intuitive to use fiducials on the prostate as > illustrated in > http://wiki.na-mic.org/Wiki/images/e/ee/KarthikMRPhantomWithEndorectalCoilBalloon.jpg > > --- > Secondly, what is the role of the "Beta" parameter in the segmentation > module. > > It would help if there was more documentation. Perhaps a dataset with a > step by step walkthrough that one could try. The only documentation that > I can find on the segmentation algorithm comes from the RandomWalker > page here: > > http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation > > > Thanks again for your help > Regards > -- > karthik > > On Thu, Mar 5, 2009 at 11:39 AM, Siddharth Vikal > wrote: > > I?ve just committed the latest module version. > > > > But the module is still windows specific in some places for the > reasons mentioned. > > > > Thanks > > Siddharth > > > > *From:* Karthik Krishnan [mailto:karthik.krishnan at kitware.com > ] > *Sent:* Thursday, March 05, 2009 11:01 AM > > *To:* Siddharth Vikal > *Cc:* Steve Pieper; slicer-devel at bwh.harvard.edu > ; Yi Gao > - Show quoted text - > *Subject:* Re: [slicer-devel] Prostate segmentation > [TRProstateBiopsy module] build issues > > - Show quoted text - > > > > Perfect. I'll wait for your heads up. > > Thanks > -- > karthik > > On Thu, Mar 5, 2009 at 10:58 AM, Siddharth Vikal > > wrote: > > Hi Karthik, > > > > You are right. As of now, the module is actually windows specific. > The module?s objective is integrate with the MRI-compatible robotic > device developed at Johns Hopkins. That robot has sensors, and we > have only windows specific drivers for them. You can find that > specific code in this file: vtkTRProstateBiopsyUSBOpticalEncoder.h, .cxx > > > > I?m going to commit latest code in 15-20 minutes, and I?ll give you > a heads up. > > > > Thanks > > Siddharth > > > > *From:* Karthik Krishnan [mailto:karthik.krishnan at kitware.com > ] > *Sent:* Thursday, March 05, 2009 10:36 AM > *To:* Siddharth Vikal > *Cc:* Steve Pieper; slicer-devel at bwh.harvard.edu > ; Yi Gao > > - Show quoted text - > > > *Subject:* Re: [slicer-devel] Prostate segmentation > [TRProstateBiopsy module] build issues > > - Show quoted text - > > > > Siddharth: > > Thanks for the quick response. > > In trying to build the module from its new location, it appears that > it will build only on windows. I get build errors on my linux > machine such as : > > In file included from > /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.cxx:1: > /home/karthik/NAMIC/NAMICSandBox/Queens/TRProstateBiopsy/vtkTRProstateBiopsySecondaryMonitor.h:7:21: > error: Windows.h: No such file or directory > > > Upon closer examination of the code, it doesn't appear that there is > anything windows specific in the workflow/functionality itself. > However, the code has been written to includet he > vtkWin32OpenGLRenderWindow, instead of using the vtkRenderWindow and > taking advantage of VTK's abstraction. > > Could you please tell us if the module windows specific ? > > Thanks > -- > karthik > > On Thu, Mar 5, 2009 at 10:17 AM, Siddharth Vikal > > wrote: > > Hi Karthik, > > After reading the emails below, I did svn diff of CMakeLists.txt > (working base vs working copy). And all the include directories are > there! > > Then, I looked closer in the email exchange, and it comes to my > attention that you are looking at the wrong location for the module. > > The correct location is here: > > http://svn.na-mic.org/NAMICSandBox/trunk/Queens/TRProstateBiopsy/ > > instead of, as you quote. > > > >> In file included from > > > >> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > > >> > > I believe the above location is very very very old, which was > created by David Gobbi over an year ago. The module was moved to > /Queens/TRProstateBiopsy in June 2008 project week, and we have been > committing to that location. > > I would request you not to waste further time to correct errors, but > to access the module from the above new location. I have added this > location the project page on wiki: > > http://www.na-mic.org/Wiki/index.php/DBP2:JHU:Roadmap > > I?ll commit the latest version right now. > > Let me know if there are queries or if run into problems this time. > > Best regards > > Siddharth > > - Show quoted text - > > -----Original Message----- > From: gaoyi.cn @gmail.com > [mailto:gaoyi.cn at gmail.com] On Behalf Of Yi Gao > Sent: Thursday, March 05, 2009 9:00 AM > To: Karthik Krishnan > Cc: Steve Pieper; slicer-devel at bwh.harvard.edu > ; Siddharth Vikal > Subject: Re: [slicer-devel] Prostate segmentation [TRProstateBiopsy > module] build issues > > Hi Karthik, > > I'm sorry that I knew little about this module.... but I think > > Siddharth (CCed) may help on that. > > Best, > > yi > > On Wed, Mar 4, 2009 at 9:06 PM, Karthik Krishnan > > > > wrote: > > > Thanks Steve. That's what I've been doing :) Adding include > directories. > > > Still get a host of build errors. I'll post more results after > poking around > > > soon. > > > > > > Thanks > > > Regards > > > -- > > > karthik > > > - Show quoted text - > > > > > > > > > On Wed, Mar 4, 2009 at 7:17 PM, Steve Pieper > > wrote: > > > > > > > > Hi Karthik - > > > > > > > > I don't know much about that module, but from the build error > looks like > > > > you may need to add the volumes module to your > include_directories command > > > > (these moved a while ago but maybe this module was never updated). > > > > > > > > include_directories( > > > > .... > > > > ${Volumes_SOURCE_DIR} > > > > ${Volumes_BINARY_DIR} > > > > ) > > > > > > > > Good luck, > > > > Steve > > > > > > > > Karthik Krishnan wrote: > > > >> > > > >> Hello: > > > >> > > > >> Is the following module available for use/testing in the Slicer > > > >> repository, NAMIC Sandbox or elsewhere ? > > > >> > > > >> http://www.na-mic.org/Wiki/index.php/Projects:ProstateSegmentation > > > >> > > > >> > > > >> I get a build errors building the TRProstateBiopsy module in the > > > >> NAMICSandbox : > > > >> > > > >> In file included from > > > >> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.cxx:18: > > > >> > > > >> > /home/karthik/NAMIC/NAMICSandBox/IGTLoadableModules/TRProstateBiopsy/vtkTRProstateBiopsyLogic.h:28:35: > > > >> error: vtkSlicerVolumesLogic.h: No such file or directory > > > >> > > > >> Not all directories in Slicer are provided as include > directories for > > > >> this module, which makes me think that there must be some other > recommended > > > >> way of building this module. > > > >> > > > >> I just tried building > NAMICSandBox/IGTLoadableModules/TRProstateBiopsy, > > > >> setting Slicer3_DIR to my build of Slicer3. > > > >> > > > >> Any help would be appreciated. > > > >> > > > >> Thanks > > > >> Regards > > > >> > > > >> -- > > > >> Karthik Krishnan > > > >> R&D Engineer, > > > >> Kitware Inc. > > > >> Ph: 518 371 3971 x119 > > > >> Fax: 518 371 3971 > > > >> > > > >> > > > >> > ------------------------------------------------------------------------ > > > >> > > > >> _______________________________________________ > > > >> slicer-devel mailing list > > > >> slicer-devel at bwh.harvard.edu > > > >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > > >> To unsubscribe: send email to > > > >> slicer-devel-request at massmail.spl.harvard.edu > with > unsubscribe as the > > > >> subject > > > > > > > > > > > > -- > > > Karthik Krishnan > > > R&D Engineer, > > > Kitware Inc. > > > Ph: 518 371 3971 x119 > > > Fax: 518 371 3971 > > > > > > > > -- > Karthik Krishnan > R&D Engineer, > Kitware Inc. > Ph: 518 371 3971 x119 > Fax: 518 371 3971 > > > > > -- > Karthik Krishnan > R&D Engineer, > Kitware Inc. > Ph: 518 371 3971 x119 > Fax: 518 371 3971 > > > > > -- > Karthik Krishnan > R&D Engineer, > Kitware Inc. > Ph: 518 371 3971 x119 > Fax: 518 371 3971 > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From fedorov at bwh.harvard.edu Thu Mar 5 15:27:57 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Thu, 5 Mar 2009 15:27:57 -0500 Subject: [slicer-devel] ITK wrappings in Slicer Message-ID: <82f536810903051227u198ad33dx4255b6594021fd5f@mail.gmail.com> Hi, What is the status of ITK wrapping in Python? I see some notes on Slicer wiki, but it is not clear to me what is available, and what is planned to be available. I would like to do some prototyping that will need composition of different ITK filters and playing with parameters, and I just wanted to know what is the least painful way to do this in Slicer. Also, is ITK wrapping available for Tcl? I hope these are quick questions to answer. Thank you Andriy Fedorov From Stephen.Aylward at Kitware.com Thu Mar 5 16:36:58 2009 From: Stephen.Aylward at Kitware.com (Stephen Aylward) Date: Thu, 5 Mar 2009 16:36:58 -0500 Subject: [slicer-devel] [Fwd: Rigid registration of aligned images] In-Reply-To: <49AFE11B.9030209@bwh.harvard.edu> References: <49AFE11B.9030209@bwh.harvard.edu> Message-ID: Hi, Works fine using the RegisterImages module in Slicer. RegisterImages croppedLung.nrrd GaussianBlurCroppedLung.nrrd --registration PipelineRigid --resampledImage out.mha I tested your data using every one of the registration forms that RegisterImages supports (i.e., PipelineRigid, Rigid, PipelineAffine, Affine, and BSpline), and it worked. Documentation is online at: http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization#Slicer3_Module:_RegisterImages Stephen On Thu, Mar 5, 2009 at 9:26 AM, Ron Kikinis wrote: > > > -------- Original Message -------- > Subject: [slicer-devel] Rigid registration of aligned images > Date: Thu, 5 Mar 2009 09:23:57 -0500 > From: Andriy Fedorov > To: Slicer Development > CC: Miller, James V (GE, Research) > > Hi all, > > This is in relation to the previous post from Yanling > http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html > > I attach the images Yanling sent to me, which create problems during > rigid registration. Note, the second image is simply the first image > after Gaussian blur. Both images have the same size, origin, > orientation and spacing. > > If I use the default registration parameters, it fails with "ERROR: > MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF > summed to zero" > > I tried to choose the parameters to adapt to the input data: I reduced > the number of samples to the minimum possible (1000), since the images > are relatively small (81x77x11), reduced the translation scaling and > number of iterations. Here are the parameters: > > Histogram bins: 30 > Spatial samples: 1000 > Iterations: 100,100,50,20 > Learning rates: default settings > Translation scaling: 10 > > Now registration fails with the message "ERROR: > MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many > samples map outside moving image buffer: 231 / 1000" > > How is it possible, if the images seemingly have 100% overlap? > > I tried different combinations of parameters, but none seems to work > for these images. > > Can anyone explain what is going on? Jim, can you help? > > Thank you > > Andriy Fedorov > > > > On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov > wrote: >> >> Yanling, >> >> The error probably was "ERROR: Failed to align input scans!", not "can >> not be aligned". >> >> This error is a generic error that means rigid registration failed for >> one or another reason. This is explained in the ChangeTracker >> documentation: >> >> >> http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 >> >> The parameters for registration in ChangeTracker were tuned for the >> meningioma brain MRI, and if registration fails, there is no mechanism >> to re-adjust the parameters. I understand it should not fail, because >> images are the same, but that is what I believe is happening. >> >> Can you try to run RigidRegistration module with the default >> parameters on that data you have and see if it completes without >> errors? >> >> Please let me know what happens if you try what I suggest. >> >> Andriy Fedorov >> >> >> >> On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: >>> >>> Hello, when testing the ChangeTracker module with two input volumes I get >>> an >>> error message says "input scans can not be aligned". The first volume is >>> the >>> original scan with spherical lung tumor. In the second volume I applied >>> Gaussian filter to grow the tumor a little bit. These two volumes ?has >>> same >>> parameters such as dimension, origin, position and spacings so the error >>> message confuses me.... >>> >>> Thanks, >>> Yanling >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject >>> >> > > > -- > Ron Kikinis, M.D., > Professor of Radiology, Harvard Medical School > Director, Surgical Planning Laboratory > http://www.spl.harvard.edu/~kikinis > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > > -- Stephen R. Aylward, Ph.D. Chief Medical Scientist Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 From jvs at bwh.harvard.edu Thu Mar 5 16:40:33 2009 From: jvs at bwh.harvard.edu (Julien von Siebenthal) Date: Thu, 05 Mar 2009 16:40:33 -0500 Subject: [slicer-devel] ITK wrappings in Slicer In-Reply-To: <82f536810903051227u198ad33dx4255b6594021fd5f@mail.gmail.com> References: <82f536810903051227u198ad33dx4255b6594021fd5f@mail.gmail.com> Message-ID: <49B046D1.3090008@bwh.harvard.edu> Hi Andriy, The solution exists. It is called WrapITK and is part of the current ITK distribution. You have however to modify the configuration to enable the wrapping in python or tcl. Be careful, you should accept most of the default settings mostly the types wrapped. It can take hours to compile the wrapping if you want to wrap all types. Best, -Julien Andriy Fedorov wrote: > Hi, > > What is the status of ITK wrapping in Python? I see some notes on > Slicer wiki, but it is not clear to me what is available, and what is > planned to be available. > > I would like to do some prototyping that will need composition of > different ITK filters and playing with parameters, and I just wanted > to know what is the least painful way to do this in Slicer. Also, is > ITK wrapping available for Tcl? > > I hope these are quick questions to answer. Thank you > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > -- Julien von Siebenthal, Ph.D. Psychiatric Neuroimaging Laboratory, Harvard Medical School Brigham and Women's Hospital, 75 Francis St., Boston MA 02115 jvs at bwh.harvard.edu From fedorov at bwh.harvard.edu Thu Mar 5 16:53:42 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Thu, 5 Mar 2009 16:53:42 -0500 Subject: [slicer-devel] [Fwd: Rigid registration of aligned images] In-Reply-To: References: <49AFE11B.9030209@bwh.harvard.edu> Message-ID: <82f536810903051353l5491086ub82d3cbcbd5efad3@mail.gmail.com> On Thu, Mar 5, 2009 at 4:36 PM, Stephen Aylward wrote: > Works fine using the RegisterImages module in Slicer. > Stephen, The problem I reported was with "RigidRegistration" module, not "RegisterImages". I am sorry if this was not clear from my post. I am using rigid registration as a processing component in a wizard pipeline. In my case, registration is supposed to be happening without user guidance. Do you suggest I should use "RegisterImages" instead of "RigidRegistration"? Is one functionally better than the other? It looks to me both of the modules are based on the same ITK core functionality. Thanks AF > RegisterImages croppedLung.nrrd GaussianBlurCroppedLung.nrrd > --registration PipelineRigid --resampledImage out.mha > > I tested your data using every one of the registration forms that > RegisterImages supports (i.e., PipelineRigid, Rigid, PipelineAffine, > Affine, and BSpline), and it worked. > > Documentation is online at: > http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization#Slicer3_Module:_RegisterImages > > Stephen > > On Thu, Mar 5, 2009 at 9:26 AM, Ron Kikinis wrote: >> >> >> -------- Original Message -------- >> Subject: [slicer-devel] Rigid registration of aligned images >> Date: Thu, 5 Mar 2009 09:23:57 -0500 >> From: Andriy Fedorov >> To: Slicer Development >> CC: Miller, James V (GE, Research) >> >> Hi all, >> >> This is in relation to the previous post from Yanling >> http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html >> >> I attach the images Yanling sent to me, which create problems during >> rigid registration. Note, the second image is simply the first image >> after Gaussian blur. Both images have the same size, origin, >> orientation and spacing. >> >> If I use the default registration parameters, it fails with "ERROR: >> MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF >> summed to zero" >> >> I tried to choose the parameters to adapt to the input data: I reduced >> the number of samples to the minimum possible (1000), since the images >> are relatively small (81x77x11), reduced the translation scaling and >> number of iterations. Here are the parameters: >> >> Histogram bins: 30 >> Spatial samples: 1000 >> Iterations: 100,100,50,20 >> Learning rates: default settings >> Translation scaling: 10 >> >> Now registration fails with the message "ERROR: >> MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many >> samples map outside moving image buffer: 231 / 1000" >> >> How is it possible, if the images seemingly have 100% overlap? >> >> I tried different combinations of parameters, but none seems to work >> for these images. >> >> Can anyone explain what is going on? Jim, can you help? >> >> Thank you >> >> Andriy Fedorov >> >> >> >> On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov >> wrote: >>> >>> Yanling, >>> >>> The error probably was "ERROR: Failed to align input scans!", not "can >>> not be aligned". >>> >>> This error is a generic error that means rigid registration failed for >>> one or another reason. This is explained in the ChangeTracker >>> documentation: >>> >>> >>> http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 >>> >>> The parameters for registration in ChangeTracker were tuned for the >>> meningioma brain MRI, and if registration fails, there is no mechanism >>> to re-adjust the parameters. I understand it should not fail, because >>> images are the same, but that is what I believe is happening. >>> >>> Can you try to run RigidRegistration module with the default >>> parameters on that data you have and see if it completes without >>> errors? >>> >>> Please let me know what happens if you try what I suggest. >>> >>> Andriy Fedorov >>> >>> >>> >>> On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: >>>> >>>> Hello, when testing the ChangeTracker module with two input volumes I get >>>> an >>>> error message says "input scans can not be aligned". The first volume is >>>> the >>>> original scan with spherical lung tumor. In the second volume I applied >>>> Gaussian filter to grow the tumor a little bit. These two volumes ?has >>>> same >>>> parameters such as dimension, origin, position and spacings so the error >>>> message confuses me.... >>>> >>>> Thanks, >>>> Yanling >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu >>>> with unsubscribe as the subject >>>> >>> >> >> >> -- >> Ron Kikinis, M.D., >> Professor of Radiology, Harvard Medical School >> Director, Surgical Planning Laboratory >> http://www.spl.harvard.edu/~kikinis >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> >> > > > > -- > Stephen R. Aylward, Ph.D. > Chief Medical Scientist > Kitware, Inc. - North Carolina Office > http://www.kitware.com > (518) 371-3971 x300 > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From Stephen.Aylward at Kitware.com Thu Mar 5 17:06:29 2009 From: Stephen.Aylward at Kitware.com (Stephen Aylward) Date: Thu, 5 Mar 2009 17:06:29 -0500 Subject: [slicer-devel] [Fwd: Rigid registration of aligned images] In-Reply-To: <82f536810903051353l5491086ub82d3cbcbd5efad3@mail.gmail.com> References: <49AFE11B.9030209@bwh.harvard.edu> <82f536810903051353l5491086ub82d3cbcbd5efad3@mail.gmail.com> Message-ID: Hi, Both modules have different strengths and weaknesses. Which one is "best" is application dependent. The point of my email was to offer an alternative solution that is available in Slicer and is known to work using your data. Please keep in mind that just because they are both based on ITK, it does not mean that they implement the same registration strategy, optimizers, rigid transforms (there are several rigid transforms in ITK), metrics, or parameter settings for those. I expect that speed and final results will vary significantly. s On Thu, Mar 5, 2009 at 4:53 PM, Andriy Fedorov wrote: > On Thu, Mar 5, 2009 at 4:36 PM, Stephen Aylward > wrote: >> Works fine using the RegisterImages module in Slicer. >> > > Stephen, > > The problem I reported was with "RigidRegistration" module, not > "RegisterImages". I am sorry if this was not clear from my post. > > I am using rigid registration as a processing component in a wizard > pipeline. In my case, registration is supposed to be happening without > user guidance. Do you suggest I should use "RegisterImages" instead of > "RigidRegistration"? Is one functionally better than the other? It > looks to me both of the modules are based on the same ITK core > functionality. > > Thanks > > AF > > > > >> RegisterImages croppedLung.nrrd GaussianBlurCroppedLung.nrrd >> --registration PipelineRigid --resampledImage out.mha >> >> I tested your data using every one of the registration forms that >> RegisterImages supports (i.e., PipelineRigid, Rigid, PipelineAffine, >> Affine, and BSpline), and it worked. >> >> Documentation is online at: >> http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization#Slicer3_Module:_RegisterImages >> >> Stephen >> >> On Thu, Mar 5, 2009 at 9:26 AM, Ron Kikinis wrote: >>> >>> >>> -------- Original Message -------- >>> Subject: [slicer-devel] Rigid registration of aligned images >>> Date: Thu, 5 Mar 2009 09:23:57 -0500 >>> From: Andriy Fedorov >>> To: Slicer Development >>> CC: Miller, James V (GE, Research) >>> >>> Hi all, >>> >>> This is in relation to the previous post from Yanling >>> http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html >>> >>> I attach the images Yanling sent to me, which create problems during >>> rigid registration. Note, the second image is simply the first image >>> after Gaussian blur. Both images have the same size, origin, >>> orientation and spacing. >>> >>> If I use the default registration parameters, it fails with "ERROR: >>> MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF >>> summed to zero" >>> >>> I tried to choose the parameters to adapt to the input data: I reduced >>> the number of samples to the minimum possible (1000), since the images >>> are relatively small (81x77x11), reduced the translation scaling and >>> number of iterations. Here are the parameters: >>> >>> Histogram bins: 30 >>> Spatial samples: 1000 >>> Iterations: 100,100,50,20 >>> Learning rates: default settings >>> Translation scaling: 10 >>> >>> Now registration fails with the message "ERROR: >>> MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many >>> samples map outside moving image buffer: 231 / 1000" >>> >>> How is it possible, if the images seemingly have 100% overlap? >>> >>> I tried different combinations of parameters, but none seems to work >>> for these images. >>> >>> Can anyone explain what is going on? Jim, can you help? >>> >>> Thank you >>> >>> Andriy Fedorov >>> >>> >>> >>> On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov >>> wrote: >>>> >>>> Yanling, >>>> >>>> The error probably was "ERROR: Failed to align input scans!", not "can >>>> not be aligned". >>>> >>>> This error is a generic error that means rigid registration failed for >>>> one or another reason. This is explained in the ChangeTracker >>>> documentation: >>>> >>>> >>>> http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 >>>> >>>> The parameters for registration in ChangeTracker were tuned for the >>>> meningioma brain MRI, and if registration fails, there is no mechanism >>>> to re-adjust the parameters. I understand it should not fail, because >>>> images are the same, but that is what I believe is happening. >>>> >>>> Can you try to run RigidRegistration module with the default >>>> parameters on that data you have and see if it completes without >>>> errors? >>>> >>>> Please let me know what happens if you try what I suggest. >>>> >>>> Andriy Fedorov >>>> >>>> >>>> >>>> On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: >>>>> >>>>> Hello, when testing the ChangeTracker module with two input volumes I get >>>>> an >>>>> error message says "input scans can not be aligned". The first volume is >>>>> the >>>>> original scan with spherical lung tumor. In the second volume I applied >>>>> Gaussian filter to grow the tumor a little bit. These two volumes ?has >>>>> same >>>>> parameters such as dimension, origin, position and spacings so the error >>>>> message confuses me.... >>>>> >>>>> Thanks, >>>>> Yanling >>>>> >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu >>>>> with unsubscribe as the subject >>>>> >>>> >>> >>> >>> -- >>> Ron Kikinis, M.D., >>> Professor of Radiology, Harvard Medical School >>> Director, Surgical Planning Laboratory >>> http://www.spl.harvard.edu/~kikinis >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject >>> >>> >> >> >> >> -- >> Stephen R. Aylward, Ph.D. >> Chief Medical Scientist >> Kitware, Inc. - North Carolina Office >> http://www.kitware.com >> (518) 371-3971 x300 >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > -- Stephen R. Aylward, Ph.D. Chief Medical Scientist Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 From curtislisle at knowledgevis.com Thu Mar 5 17:22:30 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Thu, 5 Mar 2009 17:22:30 -0500 Subject: [slicer-devel] ITK wrappings in Slicer In-Reply-To: <49B046D1.3090008@bwh.harvard.edu> References: <82f536810903051227u198ad33dx4255b6594021fd5f@mail.gmail.com> <49B046D1.3090008@bwh.harvard.edu> Message-ID: <50ADAAC6-A189-4B0E-9158-AB418E1A1E07@knowledgevis.com> There are some hidden complexities here, if you want to use ITK filters in Slicer, as far as I understand. Python wrapping can be enabled using WrapITK to create python wrapping of filters. However this ITK-specific wrapping is a little different than Slicer's standard Python wrapping. You will be able to instantiate ITK filters in python scripts, but will need to either (1) use numpy arrays to pass image data back and forth between Slicer and ITK, or (2) hand- wrap the ITK filter like in Slicer/Libs/vtkITK. This is because ITK wrapping is done differently than VTK wrapping (for reasons of C++ templating, I believe). The basic python wrapping techniques for Slicer are shown here (which you are probably familiar with already): http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3::Python To exchange data, the data arrays have to be passed intentionally between the frameworks. I am still waiting to get the matching versions of ITK and numPy to make this work on my system. However, it should go something like this: from Slicer import slicer import itk vol = slicer.MRMLScene.GetNodeFromId('vtkMRMLScalarVolumeNode1') arr = vol.GetImageData().ToArray() itkimage = itk.PyBuffer.IUS2.GetImageFromArray(arr) ... your ITK code here ... outarr = itk.PyBuffer.IUS2.GetArrayFromImage(itkimage) vol.GetImageData().FromArray(outarr) On Mar 5, 2009, at 4:40 PM, Julien von Siebenthal wrote: > Hi Andriy, > > The solution exists. It is called WrapITK and is part of the current > ITK > distribution. You have however to modify the configuration to enable > the > wrapping in python or tcl. Be careful, you should accept most of the > default settings mostly the types wrapped. It can take hours to > compile > the wrapping if you want to wrap all types. > > Best, > -Julien > > Andriy Fedorov wrote: >> Hi, >> >> What is the status of ITK wrapping in Python? I see some notes on >> Slicer wiki, but it is not clear to me what is available, and what is >> planned to be available. >> >> I would like to do some prototyping that will need composition of >> different ITK filters and playing with parameters, and I just wanted >> to know what is the least painful way to do this in Slicer. Also, is >> ITK wrapping available for Tcl? >> >> I hope these are quick questions to answer. Thank you >> >> Andriy Fedorov >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> > > > -- > Julien von Siebenthal, Ph.D. > Psychiatric Neuroimaging Laboratory, Harvard Medical School > Brigham and Women's Hospital, 75 Francis St., Boston MA 02115 > jvs at bwh.harvard.edu > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From fedorov at bwh.harvard.edu Thu Mar 5 17:35:55 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Thu, 5 Mar 2009 17:35:55 -0500 Subject: [slicer-devel] [Fwd: Rigid registration of aligned images] In-Reply-To: References: <49AFE11B.9030209@bwh.harvard.edu> <82f536810903051353l5491086ub82d3cbcbd5efad3@mail.gmail.com> Message-ID: <82f536810903051435i2defee38q82679f3b1987b0b1@mail.gmail.com> Hi Stephen, On Thu, Mar 5, 2009 at 5:06 PM, Stephen Aylward wrote: > Both modules have different strengths and weaknesses. ?Which one is > "best" is application dependent. > Are there any guidelines about what module to choose in a particular case, or is it more trial and error process? It looks to me, RegisterImages is more intuitive to initialize, but I used RigidRegistration for historical reasons. I should re-consider past results with the RegisterImages module. > The point of my email was to offer an alternative solution that is > available in Slicer and is known to work using your data. > Yes, I understand, and I appreciate this. Thank you. The data I posted is a synthetic case, which, as it appeared to me, uncovered a non-synthetic problem. My understanding was that if we find the source of the problem for this synthetic case, we may avoid future problem for similar real data. > Please keep in mind that just because they are both based on ITK, it > does not mean that they implement the same registration strategy, > optimizers, rigid transforms (there are several rigid transforms in > ITK), metrics, or parameter settings for those. ? ?I expect that speed > and final results will vary significantly. > Yes, of course -- I agree. But the fact that they are both based on ITK means that if the problem we observe in RigidRegistration is indeed an ITK problem, sooner or later it may pop up in RegisterImages if it is using the faulty component. But it may not be ITK problem, I am not sure. Thank you for the discussion, and for the registration tools. Seriously, I can feel I am standing on the shoulders of giants. AF > s > > > > On Thu, Mar 5, 2009 at 4:53 PM, Andriy Fedorov wrote: >> On Thu, Mar 5, 2009 at 4:36 PM, Stephen Aylward >> wrote: >>> Works fine using the RegisterImages module in Slicer. >>> >> >> Stephen, >> >> The problem I reported was with "RigidRegistration" module, not >> "RegisterImages". I am sorry if this was not clear from my post. >> >> I am using rigid registration as a processing component in a wizard >> pipeline. In my case, registration is supposed to be happening without >> user guidance. Do you suggest I should use "RegisterImages" instead of >> "RigidRegistration"? Is one functionally better than the other? It >> looks to me both of the modules are based on the same ITK core >> functionality. >> >> Thanks >> >> AF >> >> >> >> >>> RegisterImages croppedLung.nrrd GaussianBlurCroppedLung.nrrd >>> --registration PipelineRigid --resampledImage out.mha >>> >>> I tested your data using every one of the registration forms that >>> RegisterImages supports (i.e., PipelineRigid, Rigid, PipelineAffine, >>> Affine, and BSpline), and it worked. >>> >>> Documentation is online at: >>> http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization#Slicer3_Module:_RegisterImages >>> >>> Stephen >>> >>> On Thu, Mar 5, 2009 at 9:26 AM, Ron Kikinis wrote: >>>> >>>> >>>> -------- Original Message -------- >>>> Subject: [slicer-devel] Rigid registration of aligned images >>>> Date: Thu, 5 Mar 2009 09:23:57 -0500 >>>> From: Andriy Fedorov >>>> To: Slicer Development >>>> CC: Miller, James V (GE, Research) >>>> >>>> Hi all, >>>> >>>> This is in relation to the previous post from Yanling >>>> http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html >>>> >>>> I attach the images Yanling sent to me, which create problems during >>>> rigid registration. Note, the second image is simply the first image >>>> after Gaussian blur. Both images have the same size, origin, >>>> orientation and spacing. >>>> >>>> If I use the default registration parameters, it fails with "ERROR: >>>> MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF >>>> summed to zero" >>>> >>>> I tried to choose the parameters to adapt to the input data: I reduced >>>> the number of samples to the minimum possible (1000), since the images >>>> are relatively small (81x77x11), reduced the translation scaling and >>>> number of iterations. Here are the parameters: >>>> >>>> Histogram bins: 30 >>>> Spatial samples: 1000 >>>> Iterations: 100,100,50,20 >>>> Learning rates: default settings >>>> Translation scaling: 10 >>>> >>>> Now registration fails with the message "ERROR: >>>> MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many >>>> samples map outside moving image buffer: 231 / 1000" >>>> >>>> How is it possible, if the images seemingly have 100% overlap? >>>> >>>> I tried different combinations of parameters, but none seems to work >>>> for these images. >>>> >>>> Can anyone explain what is going on? Jim, can you help? >>>> >>>> Thank you >>>> >>>> Andriy Fedorov >>>> >>>> >>>> >>>> On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov >>>> wrote: >>>>> >>>>> Yanling, >>>>> >>>>> The error probably was "ERROR: Failed to align input scans!", not "can >>>>> not be aligned". >>>>> >>>>> This error is a generic error that means rigid registration failed for >>>>> one or another reason. This is explained in the ChangeTracker >>>>> documentation: >>>>> >>>>> >>>>> http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 >>>>> >>>>> The parameters for registration in ChangeTracker were tuned for the >>>>> meningioma brain MRI, and if registration fails, there is no mechanism >>>>> to re-adjust the parameters. I understand it should not fail, because >>>>> images are the same, but that is what I believe is happening. >>>>> >>>>> Can you try to run RigidRegistration module with the default >>>>> parameters on that data you have and see if it completes without >>>>> errors? >>>>> >>>>> Please let me know what happens if you try what I suggest. >>>>> >>>>> Andriy Fedorov >>>>> >>>>> >>>>> >>>>> On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: >>>>>> >>>>>> Hello, when testing the ChangeTracker module with two input volumes I get >>>>>> an >>>>>> error message says "input scans can not be aligned". The first volume is >>>>>> the >>>>>> original scan with spherical lung tumor. In the second volume I applied >>>>>> Gaussian filter to grow the tumor a little bit. These two volumes ?has >>>>>> same >>>>>> parameters such as dimension, origin, position and spacings so the error >>>>>> message confuses me.... >>>>>> >>>>>> Thanks, >>>>>> Yanling >>>>>> >>>>>> _______________________________________________ >>>>>> slicer-devel mailing list >>>>>> slicer-devel at bwh.harvard.edu >>>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>>> To unsubscribe: send email to >>>>>> slicer-devel-request at massmail.spl.harvard.edu >>>>>> with unsubscribe as the subject >>>>>> >>>>> >>>> >>>> >>>> -- >>>> Ron Kikinis, M.D., >>>> Professor of Radiology, Harvard Medical School >>>> Director, Surgical Planning Laboratory >>>> http://www.spl.harvard.edu/~kikinis >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>>> with unsubscribe as the subject >>>> >>>> >>> >>> >>> >>> -- >>> Stephen R. Aylward, Ph.D. >>> Chief Medical Scientist >>> Kitware, Inc. - North Carolina Office >>> http://www.kitware.com >>> (518) 371-3971 x300 >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >> > > > > -- > Stephen R. Aylward, Ph.D. > Chief Medical Scientist > Kitware, Inc. - North Carolina Office > http://www.kitware.com > (518) 371-3971 x300 > From bill.lorensen at gmail.com Thu Mar 5 17:57:07 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Thu, 5 Mar 2009 14:57:07 -0800 Subject: [slicer-devel] [Fwd: Rigid registration of aligned images] In-Reply-To: <82f536810903051435i2defee38q82679f3b1987b0b1@mail.gmail.com> References: <49AFE11B.9030209@bwh.harvard.edu> <82f536810903051353l5491086ub82d3cbcbd5efad3@mail.gmail.com> <82f536810903051435i2defee38q82679f3b1987b0b1@mail.gmail.com> Message-ID: <4db4735c0903051457g312074c5u36610fc47583d1af@mail.gmail.com> RigidRigistration is an early command line module with a simple interface. It was meant to be more of an example rather than a robust registration module. If RegisterImages is more robust, I recommend switching to it. In the future we could probably reimplement RigidRegistration to use RegisterImages internally. Bill On Thu, Mar 5, 2009 at 2:35 PM, Andriy Fedorov wrote: > Hi Stephen, > > On Thu, Mar 5, 2009 at 5:06 PM, Stephen Aylward > wrote: >> Both modules have different strengths and weaknesses. ?Which one is >> "best" is application dependent. >> > > Are there any guidelines about what module to choose in a particular > case, or is it more trial and error process? > > It looks to me, RegisterImages is more intuitive to initialize, but I > used RigidRegistration for historical reasons. I should re-consider > past results with the RegisterImages module. > >> The point of my email was to offer an alternative solution that is >> available in Slicer and is known to work using your data. >> > > Yes, I understand, and I appreciate this. Thank you. > > The data I posted is a synthetic case, which, as it appeared to me, > uncovered a non-synthetic problem. My understanding was that if we > find the source of the problem for this synthetic case, we may avoid > future problem for similar real data. > >> Please keep in mind that just because they are both based on ITK, it >> does not mean that they implement the same registration strategy, >> optimizers, rigid transforms (there are several rigid transforms in >> ITK), metrics, or parameter settings for those. ? ?I expect that speed >> and final results will vary significantly. >> > > Yes, of course -- I agree. But the fact that they are both based on > ITK means that if the problem we observe in RigidRegistration is > indeed an ITK problem, sooner or later it may pop up in RegisterImages > if it is using the faulty component. But it may not be ITK problem, I > am not sure. > > Thank you for the discussion, and for the registration tools. > Seriously, I can feel I am standing on the shoulders of giants. > > AF > > > >> s >> >> >> >> On Thu, Mar 5, 2009 at 4:53 PM, Andriy Fedorov wrote: >>> On Thu, Mar 5, 2009 at 4:36 PM, Stephen Aylward >>> wrote: >>>> Works fine using the RegisterImages module in Slicer. >>>> >>> >>> Stephen, >>> >>> The problem I reported was with "RigidRegistration" module, not >>> "RegisterImages". I am sorry if this was not clear from my post. >>> >>> I am using rigid registration as a processing component in a wizard >>> pipeline. In my case, registration is supposed to be happening without >>> user guidance. Do you suggest I should use "RegisterImages" instead of >>> "RigidRegistration"? Is one functionally better than the other? It >>> looks to me both of the modules are based on the same ITK core >>> functionality. >>> >>> Thanks >>> >>> AF >>> >>> >>> >>> >>>> RegisterImages croppedLung.nrrd GaussianBlurCroppedLung.nrrd >>>> --registration PipelineRigid --resampledImage out.mha >>>> >>>> I tested your data using every one of the registration forms that >>>> RegisterImages supports (i.e., PipelineRigid, Rigid, PipelineAffine, >>>> Affine, and BSpline), and it worked. >>>> >>>> Documentation is online at: >>>> http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization#Slicer3_Module:_RegisterImages >>>> >>>> Stephen >>>> >>>> On Thu, Mar 5, 2009 at 9:26 AM, Ron Kikinis wrote: >>>>> >>>>> >>>>> -------- Original Message -------- >>>>> Subject: [slicer-devel] Rigid registration of aligned images >>>>> Date: Thu, 5 Mar 2009 09:23:57 -0500 >>>>> From: Andriy Fedorov >>>>> To: Slicer Development >>>>> CC: Miller, James V (GE, Research) >>>>> >>>>> Hi all, >>>>> >>>>> This is in relation to the previous post from Yanling >>>>> http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html >>>>> >>>>> I attach the images Yanling sent to me, which create problems during >>>>> rigid registration. Note, the second image is simply the first image >>>>> after Gaussian blur. Both images have the same size, origin, >>>>> orientation and spacing. >>>>> >>>>> If I use the default registration parameters, it fails with "ERROR: >>>>> MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF >>>>> summed to zero" >>>>> >>>>> I tried to choose the parameters to adapt to the input data: I reduced >>>>> the number of samples to the minimum possible (1000), since the images >>>>> are relatively small (81x77x11), reduced the translation scaling and >>>>> number of iterations. Here are the parameters: >>>>> >>>>> Histogram bins: 30 >>>>> Spatial samples: 1000 >>>>> Iterations: 100,100,50,20 >>>>> Learning rates: default settings >>>>> Translation scaling: 10 >>>>> >>>>> Now registration fails with the message "ERROR: >>>>> MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many >>>>> samples map outside moving image buffer: 231 / 1000" >>>>> >>>>> How is it possible, if the images seemingly have 100% overlap? >>>>> >>>>> I tried different combinations of parameters, but none seems to work >>>>> for these images. >>>>> >>>>> Can anyone explain what is going on? Jim, can you help? >>>>> >>>>> Thank you >>>>> >>>>> Andriy Fedorov >>>>> >>>>> >>>>> >>>>> On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov >>>>> wrote: >>>>>> >>>>>> Yanling, >>>>>> >>>>>> The error probably was "ERROR: Failed to align input scans!", not "can >>>>>> not be aligned". >>>>>> >>>>>> This error is a generic error that means rigid registration failed for >>>>>> one or another reason. This is explained in the ChangeTracker >>>>>> documentation: >>>>>> >>>>>> >>>>>> http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 >>>>>> >>>>>> The parameters for registration in ChangeTracker were tuned for the >>>>>> meningioma brain MRI, and if registration fails, there is no mechanism >>>>>> to re-adjust the parameters. I understand it should not fail, because >>>>>> images are the same, but that is what I believe is happening. >>>>>> >>>>>> Can you try to run RigidRegistration module with the default >>>>>> parameters on that data you have and see if it completes without >>>>>> errors? >>>>>> >>>>>> Please let me know what happens if you try what I suggest. >>>>>> >>>>>> Andriy Fedorov >>>>>> >>>>>> >>>>>> >>>>>> On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: >>>>>>> >>>>>>> Hello, when testing the ChangeTracker module with two input volumes I get >>>>>>> an >>>>>>> error message says "input scans can not be aligned". The first volume is >>>>>>> the >>>>>>> original scan with spherical lung tumor. In the second volume I applied >>>>>>> Gaussian filter to grow the tumor a little bit. These two volumes ?has >>>>>>> same >>>>>>> parameters such as dimension, origin, position and spacings so the error >>>>>>> message confuses me.... >>>>>>> >>>>>>> Thanks, >>>>>>> Yanling >>>>>>> >>>>>>> _______________________________________________ >>>>>>> slicer-devel mailing list >>>>>>> slicer-devel at bwh.harvard.edu >>>>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>>>> To unsubscribe: send email to >>>>>>> slicer-devel-request at massmail.spl.harvard.edu >>>>>>> with unsubscribe as the subject >>>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> Ron Kikinis, M.D., >>>>> Professor of Radiology, Harvard Medical School >>>>> Director, Surgical Planning Laboratory >>>>> http://www.spl.harvard.edu/~kikinis >>>>> >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>>>> with unsubscribe as the subject >>>>> >>>>> >>>> >>>> >>>> >>>> -- >>>> Stephen R. Aylward, Ph.D. >>>> Chief Medical Scientist >>>> Kitware, Inc. - North Carolina Office >>>> http://www.kitware.com >>>> (518) 371-3971 x300 >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >>>> >>> >> >> >> >> -- >> Stephen R. Aylward, Ph.D. >> Chief Medical Scientist >> Kitware, Inc. - North Carolina Office >> http://www.kitware.com >> (518) 371-3971 x300 >> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From bill.lorensen at gmail.com Thu Mar 5 18:00:01 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Thu, 5 Mar 2009 15:00:01 -0800 Subject: [slicer-devel] ITK wrappings in Slicer In-Reply-To: <50ADAAC6-A189-4B0E-9158-AB418E1A1E07@knowledgevis.com> References: <82f536810903051227u198ad33dx4255b6594021fd5f@mail.gmail.com> <49B046D1.3090008@bwh.harvard.edu> <50ADAAC6-A189-4B0E-9158-AB418E1A1E07@knowledgevis.com> Message-ID: <4db4735c0903051500w40e11363n1bd1f4c78eae5e51@mail.gmail.com> IMHO, I would stay away from ITK wrapping in Slicer3. You are better off using the vtkITK methods or writing your own vtk class that use itk internally. As Curtis pointed out, if you are mixing vtk and itk, the wrapping is not compatible. Bill On Thu, Mar 5, 2009 at 2:22 PM, Curtis Lisle wrote: > There are some hidden complexities here, if you want to use ITK > filters in Slicer, as far as I understand. ? Python wrapping can be > enabled using WrapITK to create python wrapping of filters. ?However > this ITK-specific wrapping is a little different than Slicer's > standard Python wrapping. ?You will be able to instantiate ITK filters > in python scripts, but will need to either (1) use numpy arrays to > pass image data back and forth between Slicer and ITK, or (2) hand- > wrap the ITK filter like in Slicer/Libs/vtkITK. ? ?This is because ITK > wrapping is done differently than VTK wrapping (for reasons of C++ > templating, I believe). > > The basic python wrapping techniques for Slicer are shown here (which > you are probably familiar with already): > > http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3::Python > > To exchange data, the data arrays have to be passed intentionally > between the frameworks. ?I am still waiting to get the matching > versions of ITK and numPy to make this work on my system. ?However, it > should go something like this: > > from Slicer import slicer > import itk > > vol = slicer.MRMLScene.GetNodeFromId('vtkMRMLScalarVolumeNode1') > arr = vol.GetImageData().ToArray() > itkimage = itk.PyBuffer.IUS2.GetImageFromArray(arr) > > ... your ITK code here ... > > outarr = itk.PyBuffer.IUS2.GetArrayFromImage(itkimage) > vol.GetImageData().FromArray(outarr) > > > > On Mar 5, 2009, at 4:40 PM, Julien von Siebenthal wrote: > >> Hi Andriy, >> >> The solution exists. It is called WrapITK and is part of the current >> ITK >> distribution. You have however to modify the configuration to enable >> the >> wrapping in python or tcl. Be careful, you should accept most of the >> default settings mostly the types wrapped. It can take hours to >> compile >> the wrapping if you want to wrap all types. >> >> Best, >> -Julien >> >> Andriy Fedorov wrote: >>> Hi, >>> >>> What is the status of ITK wrapping in Python? I see some notes on >>> Slicer wiki, but it is not clear to me what is available, and what is >>> planned to be available. >>> >>> I would like to do some prototyping that will need composition of >>> different ITK filters and playing with parameters, and I just wanted >>> to know what is the least painful way to do this in Slicer. Also, is >>> ITK wrapping available for Tcl? >>> >>> I hope these are quick questions to answer. Thank you >>> >>> Andriy Fedorov >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> ?with unsubscribe as the subject >>> >> >> >> -- >> Julien von Siebenthal, Ph.D. >> Psychiatric Neuroimaging Laboratory, Harvard Medical School >> Brigham and Women's Hospital, 75 Francis St., Boston MA 02115 >> jvs at bwh.harvard.edu >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> ?with unsubscribe as the subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From pieper at bwh.harvard.edu Fri Mar 6 09:39:23 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 06 Mar 2009 09:39:23 -0500 Subject: [slicer-devel] re-orienting planes In-Reply-To: <200903051052.23329.simon@math.uh.edu> References: <200903031439.57712.simon@math.uh.edu> <49AF0626.5030708@bwh.harvard.edu> <200903051052.23329.simon@math.uh.edu> Message-ID: <49B1359B.2090101@bwh.harvard.edu> Hi Simon - Thanks for the detailed example - that's exactly what's needed. I'll try to have a close look in the next few days when I have a minute. At first glance, I'm wondering if it's related to this issue: http://sourceforge.net/tracker/index.php?func=detail&aid=1899040&group_id=12997&atid=112997 There is some code in the slice controllers that tried to compensate for the limitation of the underlying scale widget and it may cause changes to the position of the slice. Or it could be something else... -Steve Simon K Alexander wrote: > Hi Steve, > > Well, I've narrowed it down quite a bit. There were two things going on. As I noted in the previous reply, the example > I sent to the list did not exhibit the rotation of the plane I had mentioned (bad bug report!), only translations of > the point p. > > So I sorted out that this rotation was my fault, it only happened in a code path that was erroneously failing to keep > XYToSlice strictly diagonal in the upper left 3x3 matrix, so when slicer code corrected this somewhere, it rotated the > plane. In many cases this was nearly correct anyway, so the effect was subtle. > > However, this has nothing to do with the main problem I was having which is that the plane would "jump" sometimes. > > It turns out this is another instance of bug 0000410 I had reported in Mantis. The spurious rotations due to my error > were throwing me off, so I didn't see it at first. > > I've attached a complete test case in "debug.zip" > > If you load the .mrml file, fire up the TCL interactor and "source debug.tcl" it will set everything up. You'll see > that the last line of debug.tcl is > > rotations::rotate-slicenumber 0 -0.1 $p0 $p1 > > > If you repeat this line in the interactor, you'll see the plane rotate around the indicated axis. However, note as you > do it that the slice planes slider is always moving to the right in its range. > > After five repetitions (six including the one called in debug.tcl) the slider will be on the far right of the range. > > The next repetition causes the plane to translate to fit into its newly constrained range. You'll note at this point > moving the slider across its entire range only moves the slice a small fraction of the total volume. > > > So it looks like I've got the bug pinned down. Do you have a good idea where the code causing bug 0000410 is located? > Can I help sort this out if you point me at the right place? This bug is really holding me back at the moment, but I > feel like it might be a lot of code to read to locate it. I'll try anyway as time allows, but any assistance is much > appreciated. > > cheers, > Simon > > > On Wednesday 04 March 2009 04:52:22 pm Steve Pieper wrote: >> Hi Simon - >> >> I think I see what's going on but I'm having a little trouble following >> the steps. Do you think you could provide a short tcl script that >> reproduces the issue? (perhaps an excerpt of what you are using?). >> >> It should look something like this: >> >> set sliceNode [$::slicer3::MRMLScene GetNthNodeByClass 0 >> "vtkMRMLSliceNode"] $sliceNode GetSliceToRAS >> [$sliceNode GetSliceToRAS] Identity >> $sliceNode SetSliceToRASByNTP >> $sliceNode UpdateMatrices >> ... (further manipulate sliceNode...) >> >> Thanks, >> Steve >> > >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject > From millerjv at gmail.com Sat Mar 7 11:48:27 2009 From: millerjv at gmail.com (millerjv) Date: Sat, 7 Mar 2009 11:48:27 -0500 Subject: [slicer-devel] [Fwd: Rigid registration of aligned images] In-Reply-To: References: <49AFE11B.9030209@bwh.harvard.edu> Message-ID: Interesting. Andriy's error is a pretty low level error reported from the MattesMI. I believe RegisterImages also uses MattesMI, so I want to rule out a bug at the level of the MI calculation. Stephen, do you have a parameter set for RegisterImages that will register fixed.nrrd and moving.nrrd in Slicer3/Libs/MRML/Testing/ TestData? I have not been able to find a parameter set that will produce a good registration of the that data with RegisterImages. It is a bit of toy example because the images are low resolution but it bothers me that I cannot find a good set of parameters. On Mar 5, 2009, at 4:36 PM, Stephen Aylward wrote: > Hi, > > Works fine using the RegisterImages module in Slicer. > > RegisterImages croppedLung.nrrd GaussianBlurCroppedLung.nrrd > --registration PipelineRigid --resampledImage out.mha > > I tested your data using every one of the registration forms that > RegisterImages supports (i.e., PipelineRigid, Rigid, PipelineAffine, > Affine, and BSpline), and it worked. > > Documentation is online at: > http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization#Slicer3_Module > :_RegisterImages > > Stephen > > On Thu, Mar 5, 2009 at 9:26 AM, Ron Kikinis > wrote: >> >> >> -------- Original Message -------- >> Subject: [slicer-devel] Rigid registration of aligned images >> Date: Thu, 5 Mar 2009 09:23:57 -0500 >> From: Andriy Fedorov >> To: Slicer Development >> CC: Miller, James V (GE, Research) >> >> Hi all, >> >> This is in relation to the previous post from Yanling >> http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html >> >> I attach the images Yanling sent to me, which create problems during >> rigid registration. Note, the second image is simply the first image >> after Gaussian blur. Both images have the same size, origin, >> orientation and spacing. >> >> If I use the default registration parameters, it fails with "ERROR: >> MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF >> summed to zero" >> >> I tried to choose the parameters to adapt to the input data: I >> reduced >> the number of samples to the minimum possible (1000), since the >> images >> are relatively small (81x77x11), reduced the translation scaling and >> number of iterations. Here are the parameters: >> >> Histogram bins: 30 >> Spatial samples: 1000 >> Iterations: 100,100,50,20 >> Learning rates: default settings >> Translation scaling: 10 >> >> Now registration fails with the message "ERROR: >> MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many >> samples map outside moving image buffer: 231 / 1000" >> >> How is it possible, if the images seemingly have 100% overlap? >> >> I tried different combinations of parameters, but none seems to work >> for these images. >> >> Can anyone explain what is going on? Jim, can you help? >> >> Thank you >> >> Andriy Fedorov >> >> >> >> On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov > > >> wrote: >>> >>> Yanling, >>> >>> The error probably was "ERROR: Failed to align input scans!", not >>> "can >>> not be aligned". >>> >>> This error is a generic error that means rigid registration failed >>> for >>> one or another reason. This is explained in the ChangeTracker >>> documentation: >>> >>> >>> http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 >>> >>> The parameters for registration in ChangeTracker were tuned for the >>> meningioma brain MRI, and if registration fails, there is no >>> mechanism >>> to re-adjust the parameters. I understand it should not fail, >>> because >>> images are the same, but that is what I believe is happening. >>> >>> Can you try to run RigidRegistration module with the default >>> parameters on that data you have and see if it completes without >>> errors? >>> >>> Please let me know what happens if you try what I suggest. >>> >>> Andriy Fedorov >>> >>> >>> >>> On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu >>> wrote: >>>> >>>> Hello, when testing the ChangeTracker module with two input >>>> volumes I get >>>> an >>>> error message says "input scans can not be aligned". The first >>>> volume is >>>> the >>>> original scan with spherical lung tumor. In the second volume I >>>> applied >>>> Gaussian filter to grow the tumor a little bit. These two >>>> volumes has >>>> same >>>> parameters such as dimension, origin, position and spacings so >>>> the error >>>> message confuses me.... >>>> >>>> Thanks, >>>> Yanling >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu >>>> with unsubscribe as the subject >>>> >>> >> >> >> -- >> Ron Kikinis, M.D., >> Professor of Radiology, Harvard Medical School >> Director, Surgical Planning Laboratory >> http://www.spl.harvard.edu/~kikinis >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> >> > > > > -- > Stephen R. Aylward, Ph.D. > Chief Medical Scientist > Kitware, Inc. - North Carolina Office > http://www.kitware.com > (518) 371-3971 x300 > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From fedorov at bwh.harvard.edu Sat Mar 7 20:26:38 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sat, 7 Mar 2009 20:26:38 -0500 Subject: [slicer-devel] Slicer tmp directory Message-ID: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> Hi, It appears that Slicer on linux is now using ~/Slicer3 as tmp directory. I think it used to be /tmp/Slicer3 in the past. I just wanted to comment that this change may create problems for users not very familiar with Linux. For example, on SPL net, when the minuscule quota is exhausted, Slicer will just hang during startup. And if the quota is exhausted while running Slicer, it will give messages like failure to read/write file in the error log. To fix this, the user needs to free space, restart Slicer, find the place in Slicer settings where the tmp directory is set, and reset it to a location without space limit. And this can happen only *if* the user understands that the problem is in insufficient disk space, because Slicer doesn't give any popup messages with the explanations. The reason I bring this up is because I just had to go through all these steps for someone working on DTI sitting next to me, because he had no clue why Slicer doesn't work. Andriy Fedorov From hayes at bwh.harvard.edu Sat Mar 7 20:34:14 2009 From: hayes at bwh.harvard.edu (Kathryn Hayes) Date: Sat, 7 Mar 2009 20:34:14 -0500 (EST) Subject: [slicer-devel] Slicer tmp directory In-Reply-To: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> Message-ID: I'd like to find a better solution to this. It was originally in /tmp, but every time someone needed to reboot, their module cache was blown away. So, I moved it to /var/tmp, but it appears that Mac blows away /var/tmp on reboot as well. People who use Mac were not pleased with how long it was taking after a reboot to start Slicer. Does anyone have a suggestion for a place to put the module cache so that the state is preserved across reboots, generic enough to be used on most systems, and compatible with the micro-sized home directories at the SPL? Thanks, Katie On Sat, 7 Mar 2009, Andriy Fedorov wrote: > Hi, > > It appears that Slicer on linux is now using ~/Slicer3 as > tmp directory. I think it used to be /tmp/Slicer3 in the > past. > > I just wanted to comment that this change may create problems for > users not very familiar with Linux. For example, on SPL net, when the > minuscule quota is exhausted, Slicer will just hang during startup. > And if the quota is exhausted while running Slicer, it will give > messages like failure to read/write file in the error log. > > To fix this, the user needs to free space, restart Slicer, find the > place in Slicer settings where the tmp directory is set, and reset it > to a location without space limit. And this can happen only *if* the > user understands that the problem is in insufficient disk space, > because Slicer doesn't give any popup messages with the explanations. > > The reason I bring this up is because I just had to go through all > these steps for someone working on DTI sitting next to me, because he > had no clue why Slicer doesn't work. > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From fedorov at bwh.harvard.edu Sat Mar 7 20:38:26 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sat, 7 Mar 2009 20:38:26 -0500 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> Message-ID: <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> On Sat, Mar 7, 2009 at 8:34 PM, Kathryn Hayes wrote: > Does anyone have a suggestion for a place to put the module cache so that > the state is preserved across reboots, generic enough to be used on most > systems, and compatible with the micro-sized home directories at the SPL? > Just one more thing to consider. tmp directory in Slicer is used not only for module cache, but also, in particular, for file IO in CLI execution. So the tmp directory should either be in a location with relatively large disk space available, or the module cache location should be separated from the CLI file IO location. > Thanks, > > Katie > > On Sat, 7 Mar 2009, Andriy Fedorov wrote: > >> Hi, >> >> It appears that Slicer on linux is now using ~/Slicer3 as >> tmp directory. I think it used to be /tmp/Slicer3 in the >> past. >> >> I just wanted to comment that this change may create problems for >> users not very familiar with Linux. For example, on SPL net, when the >> minuscule quota is exhausted, Slicer will just hang during startup. >> And if the quota is exhausted while running Slicer, it will give >> messages like failure to read/write file in the error log. >> >> To fix this, the user needs to free space, restart Slicer, find the >> place in Slicer settings where the tmp directory is set, and reset it >> to a location without space limit. And this can happen only *if* the >> user understands that the problem is in insufficient disk space, >> because Slicer doesn't give any popup messages with the explanations. >> >> The reason I bring this up is because I just had to go through all >> these steps for someone working on DTI sitting next to me, because he >> had no clue why Slicer doesn't work. >> >> Andriy Fedorov >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > From mhalle at bwh.harvard.edu Sat Mar 7 23:02:08 2009 From: mhalle at bwh.harvard.edu (Michael Halle) Date: Sat, 07 Mar 2009 23:02:08 -0500 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> Message-ID: <49B34340.8000909@bwh.harvard.edu> I think the idea of splitting the slicer cache up into a data cache and a program cache makes sense from a utility, efficiency, and practicality point of view. --Mike Andriy Fedorov wrote: > On Sat, Mar 7, 2009 at 8:34 PM, Kathryn Hayes wrote: >> Does anyone have a suggestion for a place to put the module cache so that >> the state is preserved across reboots, generic enough to be used on most >> systems, and compatible with the micro-sized home directories at the SPL? >> > > Just one more thing to consider. > > tmp directory in Slicer is used not only for module cache, but also, > in particular, for file IO in CLI execution. So the tmp directory > should either be in a location with relatively large disk space > available, or the module cache location should be separated from the > CLI file IO location. > > >> Thanks, >> >> Katie >> >> On Sat, 7 Mar 2009, Andriy Fedorov wrote: >> >>> Hi, >>> >>> It appears that Slicer on linux is now using ~/Slicer3 as >>> tmp directory. I think it used to be /tmp/Slicer3 in the >>> past. >>> >>> I just wanted to comment that this change may create problems for >>> users not very familiar with Linux. For example, on SPL net, when the >>> minuscule quota is exhausted, Slicer will just hang during startup. >>> And if the quota is exhausted while running Slicer, it will give >>> messages like failure to read/write file in the error log. >>> >>> To fix this, the user needs to free space, restart Slicer, find the >>> place in Slicer settings where the tmp directory is set, and reset it >>> to a location without space limit. And this can happen only *if* the >>> user understands that the problem is in insufficient disk space, >>> because Slicer doesn't give any popup messages with the explanations. >>> >>> The reason I bring this up is because I just had to go through all >>> these steps for someone working on DTI sitting next to me, because he >>> had no clue why Slicer doesn't work. >>> >>> Andriy Fedorov >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >>> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From Stephen.Aylward at Kitware.com Sun Mar 8 10:18:58 2009 From: Stephen.Aylward at Kitware.com (Stephen Aylward) Date: Sun, 8 Mar 2009 10:18:58 -0400 Subject: [slicer-devel] [Fwd: Rigid registration of aligned images] In-Reply-To: References: <49AFE11B.9030209@bwh.harvard.edu> Message-ID: That problem typically occurs the initial parameters values are poor (particularly center of rotation is not being set) and/or when the optimizers scales are too small (thereby allowing large jumps). It can also happen when using too few bins. A philosophy is to set scales so that a unit jump in any feature being optimized moves the image a magnitude of about 25% of the expected mis-alignment in that feature direction. I think my scale for offsets is 1/(4*pixel_spacing[i]) (thereby suggested the expected offset is 16 voxels in each dim i). Center of rotation defaults to center of the image. Expected rotation is 0.1 radians since medical scans typically don't have much rotation = scale of 2.5. Number of bins = 100 Using the one-plus-one optimizer followed by gradient descent. Mainly using default values for them. More details are in the .xml file in RegisterImages/Applications/RegisterImagesSlicer.xml If you're using an old ITK checkout, there was a bug in one-plus-one where it was using the square of the scales - thereby allowing really large jumps. We fixed that 6-9 months ago. I hope this helps, s On Sat, Mar 7, 2009 at 12:48 PM, millerjv wrote: > Interesting. Andriy's error is a pretty low level error reported from the > MattesMI. I believe RegisterImages also uses MattesMI, so I want to rule out > a bug at the level of the MI calculation. > > Stephen, do you have a parameter set for RegisterImages that will register > fixed.nrrd and moving.nrrd in Slicer3/Libs/MRML/Testing/TestData? I have not > been able to find a parameter set that will produce a good registration of > the that data with RegisterImages. ?It is a bit of toy example because the > images are low resolution but it bothers me that I cannot find a good set of > parameters. > > > On Mar 5, 2009, at 4:36 PM, Stephen Aylward wrote: > >> Hi, >> >> Works fine using the RegisterImages module in Slicer. >> >> RegisterImages croppedLung.nrrd GaussianBlurCroppedLung.nrrd >> --registration PipelineRigid --resampledImage out.mha >> >> I tested your data using every one of the registration forms that >> RegisterImages supports (i.e., PipelineRigid, Rigid, PipelineAffine, >> Affine, and BSpline), and it worked. >> >> Documentation is online at: >> >> http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization#Slicer3_Module:_RegisterImages >> >> Stephen >> >> On Thu, Mar 5, 2009 at 9:26 AM, Ron Kikinis >> wrote: >>> >>> >>> -------- Original Message -------- >>> Subject: [slicer-devel] Rigid registration of aligned images >>> Date: Thu, 5 Mar 2009 09:23:57 -0500 >>> From: Andriy Fedorov >>> To: Slicer Development >>> CC: Miller, James V (GE, Research) >>> >>> Hi all, >>> >>> This is in relation to the previous post from Yanling >>> http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002904.html >>> >>> I attach the images Yanling sent to me, which create problems during >>> rigid registration. Note, the second image is simply the first image >>> after Gaussian blur. Both images have the same size, origin, >>> orientation and spacing. >>> >>> If I use the default registration parameters, it fails with "ERROR: >>> MattesMutualInformationImageToImageMetric(0x7f7c34230120): Joint PDF >>> summed to zero" >>> >>> I tried to choose the parameters to adapt to the input data: I reduced >>> the number of samples to the minimum possible (1000), since the images >>> are relatively small (81x77x11), reduced the translation scaling and >>> number of iterations. Here are the parameters: >>> >>> Histogram bins: 30 >>> Spatial samples: 1000 >>> Iterations: 100,100,50,20 >>> Learning rates: default settings >>> Translation scaling: 10 >>> >>> Now registration fails with the message "ERROR: >>> MattesMutualInformationImageToImageMetric(0x7f7c3482fd10): Too many >>> samples map outside moving image buffer: 231 / 1000" >>> >>> How is it possible, if the images seemingly have 100% overlap? >>> >>> I tried different combinations of parameters, but none seems to work >>> for these images. >>> >>> Can anyone explain what is going on? Jim, can you help? >>> >>> Thank you >>> >>> Andriy Fedorov >>> >>> >>> >>> On Tue, Mar 3, 2009 at 1:30 PM, Andriy Fedorov >>> wrote: >>>> >>>> Yanling, >>>> >>>> The error probably was "ERROR: Failed to align input scans!", not "can >>>> not be aligned". >>>> >>>> This error is a generic error that means rigid registration failed for >>>> one or another reason. This is explained in the ChangeTracker >>>> documentation: >>>> >>>> >>>> >>>> http://wiki.slicer.org/slicerWiki/index.php/Modules:ChangeTracker-Documentation-3.4 >>>> >>>> The parameters for registration in ChangeTracker were tuned for the >>>> meningioma brain MRI, and if registration fails, there is no mechanism >>>> to re-adjust the parameters. I understand it should not fail, because >>>> images are the same, but that is what I believe is happening. >>>> >>>> Can you try to run RigidRegistration module with the default >>>> parameters on that data you have and see if it completes without >>>> errors? >>>> >>>> Please let me know what happens if you try what I suggest. >>>> >>>> Andriy Fedorov >>>> >>>> >>>> >>>> On Tue, Mar 3, 2009 at 10:51 AM, Yanling Liu wrote: >>>>> >>>>> Hello, when testing the ChangeTracker module with two input volumes I >>>>> get >>>>> an >>>>> error message says "input scans can not be aligned". The first volume >>>>> is >>>>> the >>>>> original scan with spherical lung tumor. In the second volume I applied >>>>> Gaussian filter to grow the tumor a little bit. These two volumes ?has >>>>> same >>>>> parameters such as dimension, origin, position and spacings so the >>>>> error >>>>> message confuses me.... >>>>> >>>>> Thanks, >>>>> Yanling >>>>> >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu >>>>> with unsubscribe as the subject >>>>> >>>> >>> >>> >>> -- >>> Ron Kikinis, M.D., >>> Professor of Radiology, Harvard Medical School >>> Director, Surgical Planning Laboratory >>> http://www.spl.harvard.edu/~kikinis >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject >>> >>> >> >> >> >> -- >> Stephen R. Aylward, Ph.D. >> Chief Medical Scientist >> Kitware, Inc. - North Carolina Office >> http://www.kitware.com >> (518) 371-3971 x300 >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu?with unsubscribe as the >> subject > > -- Stephen R. Aylward, Ph.D. Chief Medical Scientist Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 From demian.wassermann at sophia.inria.fr Sun Mar 8 13:51:39 2009 From: demian.wassermann at sophia.inria.fr (Demian Wassermann) Date: Sun, 8 Mar 2009 18:51:39 +0100 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: <49B34340.8000909@bwh.harvard.edu> References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> <49B34340.8000909@bwh.harvard.edu> Message-ID: I agree, in standard unix applications, the cache/configuration files usually go in a hidden folder at the user's home, usually $HOME/.Slicer3 and the data cache should go at /tmp/Slicer3-$USER or something like that Cheers Demian -- Demian Wassermann demian.wassermann at sophia.inria.fr PhD Student Odyssee Research Project INRIA Sophia-Antipolis 2004 route des lucioles - FR-06902 On Mar 8, 2009, at 5:02 AM, Michael Halle wrote: > I think the idea of splitting the slicer cache up into a data cache > and > a program cache makes sense from a utility, efficiency, and > practicality > point of view. > > --Mike > > > Andriy Fedorov wrote: >> On Sat, Mar 7, 2009 at 8:34 PM, Kathryn Hayes >> wrote: >>> Does anyone have a suggestion for a place to put the module cache >>> so that >>> the state is preserved across reboots, generic enough to be used >>> on most >>> systems, and compatible with the micro-sized home directories at >>> the SPL? >>> >> >> Just one more thing to consider. >> >> tmp directory in Slicer is used not only for module cache, but also, >> in particular, for file IO in CLI execution. So the tmp directory >> should either be in a location with relatively large disk space >> available, or the module cache location should be separated from the >> CLI file IO location. >> >> >>> Thanks, >>> >>> Katie >>> >>> On Sat, 7 Mar 2009, Andriy Fedorov wrote: >>> >>>> Hi, >>>> >>>> It appears that Slicer on linux is now using ~/Slicer3 as >>>> tmp directory. I think it used to be /tmp/Slicer3 in the >>>> past. >>>> >>>> I just wanted to comment that this change may create problems for >>>> users not very familiar with Linux. For example, on SPL net, >>>> when the >>>> minuscule quota is exhausted, Slicer will just hang during startup. >>>> And if the quota is exhausted while running Slicer, it will give >>>> messages like failure to read/write file in the error log. >>>> >>>> To fix this, the user needs to free space, restart Slicer, find the >>>> place in Slicer settings where the tmp directory is set, and >>>> reset it >>>> to a location without space limit. And this can happen only *if* >>>> the >>>> user understands that the problem is in insufficient disk space, >>>> because Slicer doesn't give any popup messages with the >>>> explanations. >>>> >>>> The reason I bring this up is because I just had to go through all >>>> these steps for someone working on DTI sitting next to me, >>>> because he >>>> had no clue why Slicer doesn't work. >>>> >>>> Andriy Fedorov >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel- >>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>>> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel- >> request at massmail.spl.harvard.edu with unsubscribe as the subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel- > request at massmail.spl.harvard.edu with unsubscribe as the subject From simon at math.uh.edu Sun Mar 8 14:08:07 2009 From: simon at math.uh.edu (Simon Alexander) Date: Sun, 8 Mar 2009 13:08:07 -0500 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> <49B34340.8000909@bwh.harvard.edu> Message-ID: For what it's worth, I agree this is the best approach. I'm not sure what is usually done on Windows. I suspect per user Plugin and/or setup script in $HOME/.Slicer3 would be useful too, I'm not sure what is available along those lines at the moment, as I've just been using a build directory. cheers, simon On Mar 8, 2009, at 12:51 PM, Demian Wassermann wrote: > I agree, in standard unix applications, the cache/configuration files > usually go in a hidden folder at the user's home, usually > $HOME/.Slicer3 and the data cache should go at /tmp/Slicer3-$USER or > something like that > > > Cheers > Demian > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > > > > On Mar 8, 2009, at 5:02 AM, Michael Halle wrote: > >> I think the idea of splitting the slicer cache up into a data cache >> and >> a program cache makes sense from a utility, efficiency, and >> practicality >> point of view. >> >> --Mike >> >> >> Andriy Fedorov wrote: >>> On Sat, Mar 7, 2009 at 8:34 PM, Kathryn Hayes >>> wrote: >>>> Does anyone have a suggestion for a place to put the module cache >>>> so that >>>> the state is preserved across reboots, generic enough to be used >>>> on most >>>> systems, and compatible with the micro-sized home directories at >>>> the SPL? >>>> >>> >>> Just one more thing to consider. >>> >>> tmp directory in Slicer is used not only for module cache, but also, >>> in particular, for file IO in CLI execution. So the tmp directory >>> should either be in a location with relatively large disk space >>> available, or the module cache location should be separated from the >>> CLI file IO location. >>> >>> >>>> Thanks, >>>> >>>> Katie >>>> >>>> On Sat, 7 Mar 2009, Andriy Fedorov wrote: >>>> >>>>> Hi, >>>>> >>>>> It appears that Slicer on linux is now using ~/Slicer3 >>>>> as >>>>> tmp directory. I think it used to be /tmp/Slicer3 in the >>>>> past. >>>>> >>>>> I just wanted to comment that this change may create problems for >>>>> users not very familiar with Linux. For example, on SPL net, >>>>> when the >>>>> minuscule quota is exhausted, Slicer will just hang during >>>>> startup. >>>>> And if the quota is exhausted while running Slicer, it will give >>>>> messages like failure to read/write file in the error log. >>>>> >>>>> To fix this, the user needs to free space, restart Slicer, find >>>>> the >>>>> place in Slicer settings where the tmp directory is set, and >>>>> reset it >>>>> to a location without space limit. And this can happen only *if* >>>>> the >>>>> user understands that the problem is in insufficient disk space, >>>>> because Slicer doesn't give any popup messages with the >>>>> explanations. >>>>> >>>>> The reason I bring this up is because I just had to go through all >>>>> these steps for someone working on DTI sitting next to me, >>>>> because he >>>>> had no clue why Slicer doesn't work. >>>>> >>>>> Andriy Fedorov >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to slicer-devel- >>>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>>>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel- >>> request at massmail.spl.harvard.edu with unsubscribe as the subject >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel- >> request at massmail.spl.harvard.edu with unsubscribe as the subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From fedorov at bwh.harvard.edu Sun Mar 8 17:11:58 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Sun, 8 Mar 2009 17:11:58 -0400 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> <49B34340.8000909@bwh.harvard.edu> Message-ID: <82f536810903081411r40678b84me503c592fee23c34@mail.gmail.com> I also like the suggestion of Demian. This makes perfect sense on Linux systems. I wanted to add that /tmp is not wiped out on reboot on all Linux systems. I think this is an SPL-specific configuration. Considering the trade-offs, as I understand, removal of module cache will simply result in a delayed startup. On the other hand, failure to create file IO cache in home directory will cause stall/failure of Slicer. So even storing all the settings in /tmp may be better, if separation of file IO and module cache is not trivial to implement. AF On Sun, Mar 8, 2009 at 2:08 PM, Simon Alexander wrote: > For what it's worth, I agree this is the best approach. > > I'm not sure what is usually done on Windows. > > I suspect per user Plugin and/or setup script in $HOME/.Slicer3 ?would > be useful too, I'm not sure what is available along those lines at the > moment, as I've just been using a build directory. > > cheers, > simon > > On Mar 8, 2009, at 12:51 PM, Demian Wassermann wrote: > >> I agree, in standard unix applications, the cache/configuration files >> usually go in a hidden folder at the user's home, usually >> $HOME/.Slicer3 and the data cache should go at /tmp/Slicer3-$USER or >> something like that >> >> >> Cheers >> Demian >> >> -- >> Demian Wassermann >> demian.wassermann at sophia.inria.fr >> PhD Student >> Odyssee Research Project >> INRIA Sophia-Antipolis >> 2004 route des lucioles - FR-06902 >> >> >> >> >> On Mar 8, 2009, at 5:02 AM, Michael Halle wrote: >> >>> I think the idea of splitting the slicer cache up into a data cache >>> and >>> a program cache makes sense from a utility, efficiency, and >>> practicality >>> point of view. >>> >>> --Mike >>> >>> >>> Andriy Fedorov wrote: >>>> On Sat, Mar 7, 2009 at 8:34 PM, Kathryn Hayes >>>> wrote: >>>>> Does anyone have a suggestion for a place to put the module cache >>>>> so that >>>>> the state is preserved across reboots, generic enough to be used >>>>> on most >>>>> systems, and compatible with the micro-sized home directories at >>>>> the SPL? >>>>> >>>> >>>> Just one more thing to consider. >>>> >>>> tmp directory in Slicer is used not only for module cache, but also, >>>> in particular, for file IO in CLI execution. So the tmp directory >>>> should either be in a location with relatively large disk space >>>> available, or the module cache location should be separated from the >>>> CLI file IO location. >>>> >>>> >>>>> Thanks, >>>>> >>>>> Katie >>>>> >>>>> On Sat, 7 Mar 2009, Andriy Fedorov wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> It appears that Slicer on linux is now using ~/Slicer3 >>>>>> as >>>>>> tmp directory. I think it used to be /tmp/Slicer3 in the >>>>>> past. >>>>>> >>>>>> I just wanted to comment that this change may create problems for >>>>>> users not very familiar with Linux. For example, on SPL net, >>>>>> when the >>>>>> minuscule quota is exhausted, Slicer will just hang during >>>>>> startup. >>>>>> And if the quota is exhausted while running Slicer, it will give >>>>>> messages like failure to read/write file in the error log. >>>>>> >>>>>> To fix this, the user needs to free space, restart Slicer, find >>>>>> the >>>>>> place in Slicer settings where the tmp directory is set, and >>>>>> reset it >>>>>> to a location without space limit. And this can happen only *if* >>>>>> the >>>>>> user understands that the problem is in insufficient disk space, >>>>>> because Slicer doesn't give any popup messages with the >>>>>> explanations. >>>>>> >>>>>> The reason I bring this up is because I just had to go through all >>>>>> these steps for someone working on DTI sitting next to me, >>>>>> because he >>>>>> had no clue why Slicer doesn't work. >>>>>> >>>>>> Andriy Fedorov >>>>>> _______________________________________________ >>>>>> slicer-devel mailing list >>>>>> slicer-devel at bwh.harvard.edu >>>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>>> To unsubscribe: send email to slicer-devel- >>>>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>>>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel- >>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel- >>> request at massmail.spl.harvard.edu with unsubscribe as the subject >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> ?with unsubscribe as the subject > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From bill.lorensen at gmail.com Sun Mar 8 18:52:48 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sun, 8 Mar 2009 15:52:48 -0700 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: <82f536810903081411r40678b84me503c592fee23c34@mail.gmail.com> References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> <49B34340.8000909@bwh.harvard.edu> <82f536810903081411r40678b84me503c592fee23c34@mail.gmail.com> Message-ID: <4db4735c0903081552j1ed051bbgf6fe2a41e6cdd9ae@mail.gmail.com> Usually, /tmp is wiped on reboot, but /usr/tmp is not. On Sun, Mar 8, 2009 at 2:11 PM, Andriy Fedorov wrote: > I also like the suggestion of Demian. This makes perfect sense on Linux systems. > > I wanted to add that /tmp is not wiped out on reboot on all Linux > systems. I think this is an SPL-specific configuration. > > Considering the trade-offs, as I understand, removal of module cache > will simply result in a delayed startup. On the other hand, failure to > create file IO cache in home directory will cause stall/failure of > Slicer. So even storing all the settings in /tmp may be better, if > separation of file IO and module cache is not trivial to implement. > > AF > > > On Sun, Mar 8, 2009 at 2:08 PM, Simon Alexander wrote: >> For what it's worth, I agree this is the best approach. >> >> I'm not sure what is usually done on Windows. >> >> I suspect per user Plugin and/or setup script in $HOME/.Slicer3 ?would >> be useful too, I'm not sure what is available along those lines at the >> moment, as I've just been using a build directory. >> >> cheers, >> simon >> >> On Mar 8, 2009, at 12:51 PM, Demian Wassermann wrote: >> >>> I agree, in standard unix applications, the cache/configuration files >>> usually go in a hidden folder at the user's home, usually >>> $HOME/.Slicer3 and the data cache should go at /tmp/Slicer3-$USER or >>> something like that >>> >>> >>> Cheers >>> Demian >>> >>> -- >>> Demian Wassermann >>> demian.wassermann at sophia.inria.fr >>> PhD Student >>> Odyssee Research Project >>> INRIA Sophia-Antipolis >>> 2004 route des lucioles - FR-06902 >>> >>> >>> >>> >>> On Mar 8, 2009, at 5:02 AM, Michael Halle wrote: >>> >>>> I think the idea of splitting the slicer cache up into a data cache >>>> and >>>> a program cache makes sense from a utility, efficiency, and >>>> practicality >>>> point of view. >>>> >>>> --Mike >>>> >>>> >>>> Andriy Fedorov wrote: >>>>> On Sat, Mar 7, 2009 at 8:34 PM, Kathryn Hayes >>>>> wrote: >>>>>> Does anyone have a suggestion for a place to put the module cache >>>>>> so that >>>>>> the state is preserved across reboots, generic enough to be used >>>>>> on most >>>>>> systems, and compatible with the micro-sized home directories at >>>>>> the SPL? >>>>>> >>>>> >>>>> Just one more thing to consider. >>>>> >>>>> tmp directory in Slicer is used not only for module cache, but also, >>>>> in particular, for file IO in CLI execution. So the tmp directory >>>>> should either be in a location with relatively large disk space >>>>> available, or the module cache location should be separated from the >>>>> CLI file IO location. >>>>> >>>>> >>>>>> Thanks, >>>>>> >>>>>> Katie >>>>>> >>>>>> On Sat, 7 Mar 2009, Andriy Fedorov wrote: >>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> It appears that Slicer on linux is now using ~/Slicer3 >>>>>>> as >>>>>>> tmp directory. I think it used to be /tmp/Slicer3 in the >>>>>>> past. >>>>>>> >>>>>>> I just wanted to comment that this change may create problems for >>>>>>> users not very familiar with Linux. For example, on SPL net, >>>>>>> when the >>>>>>> minuscule quota is exhausted, Slicer will just hang during >>>>>>> startup. >>>>>>> And if the quota is exhausted while running Slicer, it will give >>>>>>> messages like failure to read/write file in the error log. >>>>>>> >>>>>>> To fix this, the user needs to free space, restart Slicer, find >>>>>>> the >>>>>>> place in Slicer settings where the tmp directory is set, and >>>>>>> reset it >>>>>>> to a location without space limit. And this can happen only *if* >>>>>>> the >>>>>>> user understands that the problem is in insufficient disk space, >>>>>>> because Slicer doesn't give any popup messages with the >>>>>>> explanations. >>>>>>> >>>>>>> The reason I bring this up is because I just had to go through all >>>>>>> these steps for someone working on DTI sitting next to me, >>>>>>> because he >>>>>>> had no clue why Slicer doesn't work. >>>>>>> >>>>>>> Andriy Fedorov >>>>>>> _______________________________________________ >>>>>>> slicer-devel mailing list >>>>>>> slicer-devel at bwh.harvard.edu >>>>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>>>> To unsubscribe: send email to slicer-devel- >>>>>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>>>>>> >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to slicer-devel- >>>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel- >>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> ?with unsubscribe as the subject >> >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From mhalle at bwh.harvard.edu Sun Mar 8 19:12:29 2009 From: mhalle at bwh.harvard.edu (Michael Halle) Date: Sun, 08 Mar 2009 19:12:29 -0400 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: <4db4735c0903081552j1ed051bbgf6fe2a41e6cdd9ae@mail.gmail.com> References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> <49B34340.8000909@bwh.harvard.edu> <82f536810903081411r40678b84me503c592fee23c34@mail.gmail.com> <4db4735c0903081552j1ed051bbgf6fe2a41e6cdd9ae@mail.gmail.com> Message-ID: <49B450DD.9020105@bwh.harvard.edu> Bill Lorensen wrote: > Usually, /tmp is wiped on reboot, but /usr/tmp is not. Yes, but /usr/tmp became /var/tmp in modern UNIX systems. The Filesystem Hierarchy Standard (FHS) defines the UNIX filesystem conventions and rationales: http://www.pathname.com/fhs/pub/fhs-2.3.html http://en.wikipedia.org/wiki/Filesystem_Hierarchy_Standard Of note: * "Programs must not assume that any files or directories in /tmp are preserved between invocations of the program." * "The /var/tmp directory is made available for programs that require temporary files or directories that are preserved between system reboots. Therefore, data stored in /var/tmp is more persistent than data in /tmp. Files and directories located in /var/tmp must not be deleted when the system is booted. Although data stored in /var/tmp is typically deleted in a site-specific manner, it is recommended that deletions occur at a less frequent interval than /tmp." --Mike From bill.lorensen at gmail.com Sun Mar 8 20:35:29 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Sun, 8 Mar 2009 17:35:29 -0700 Subject: [slicer-devel] Slicer tmp directory In-Reply-To: <49B450DD.9020105@bwh.harvard.edu> References: <82f536810903071726t7006433sa8a577ee494aec4@mail.gmail.com> <82f536810903071738k2171ec1erc17ef594311a5b3e@mail.gmail.com> <49B34340.8000909@bwh.harvard.edu> <82f536810903081411r40678b84me503c592fee23c34@mail.gmail.com> <4db4735c0903081552j1ed051bbgf6fe2a41e6cdd9ae@mail.gmail.com> <49B450DD.9020105@bwh.harvard.edu> Message-ID: <4db4735c0903081735w7a66cae0v8c55b6df2256cea8@mail.gmail.com> It hard not being modern any more... On Sun, Mar 8, 2009 at 4:12 PM, Michael Halle wrote: > Bill Lorensen wrote: >> >> Usually, /tmp is wiped on reboot, but /usr/tmp is not. > > Yes, but /usr/tmp became /var/tmp in modern UNIX systems. ?The Filesystem > Hierarchy Standard (FHS) defines the UNIX filesystem conventions and > rationales: > > http://www.pathname.com/fhs/pub/fhs-2.3.html > http://en.wikipedia.org/wiki/Filesystem_Hierarchy_Standard > > Of note: > > * "Programs must not assume that any files or directories in /tmp are > preserved between invocations of the program." > > * "The /var/tmp directory is made available for programs that require > temporary files or directories that are preserved between system reboots. > Therefore, data stored in /var/tmp is more persistent than data in /tmp. > > Files and directories located in /var/tmp must not be deleted when the > system is booted. Although data stored in /var/tmp is typically deleted in a > site-specific manner, it is recommended that deletions occur at a less > frequent interval than /tmp." > > --Mike > From jross at bwh.harvard.edu Mon Mar 9 18:53:57 2009 From: jross at bwh.harvard.edu (James Ross) Date: Mon, 09 Mar 2009 18:53:57 -0400 Subject: [slicer-devel] Question re. node configuration to display label map data Message-ID: <49B59E05.4060203@bwh.harvard.edu> Greetings, I am working on a module that produces a label map resulting from a segmentation routine. I would like to strip down the GUI as much as possible for a specific usage of the module, so I am creating a vtkMRMLScalarVolumeNode "behind the scenes" (not using a selector), adding the output segmentation data to it, and then adding the node to the scene. However, while I see the output node in the MRML tree, the label map is not rendering. Mousing over different sections of the input (CT) data, I do see the correct label value indicated, so I believe that the data has been added to the node, it's just not being rendered. Relevant code: void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long event, void *callData ) { ... this->OutputNode->SetName( "MyModuleNode" ); this->OutputNode->SetLabelMap( 1 ); this->OutputNode->SetScene( this->GetMRMLScene() ); this->Logic->Apply( ..., this->OutputNode ); // See below ... this->GetMRMLScene()->AddNode( this->OutputNode ); ... } void vtkMyModuleLogic::Apply( ..., vtkMRMLScalarVolumeNode* outputNode ) { ... outputNode->SetAndObserveDisplayNodeID( inputNode->GetDisplayNodeID() ); // inputNode is a label map outputNode->CopyOrientation( inputNode ); outputNode->SetAndObserveTransformNodeID( inputNode->GetTransformNodeID() ); outputNode->SetAndObserveImageData( image ); // image deep copies filter output outputNode->SetModifiedSinceRead( 1 ); ... } I also tried creating and setting a separate display node for the OutputNode, but that didn't seem to work either. Any ideas would be greatly appreciated. Regards, -James From nicole at bwh.harvard.edu Mon Mar 9 19:13:49 2009 From: nicole at bwh.harvard.edu (Nicole Aucoin) Date: Mon, 9 Mar 2009 19:13:49 -0400 (EDT) Subject: [slicer-devel] Question re. node configuration to display label map data In-Reply-To: <49B59E05.4060203@bwh.harvard.edu> References: <49B59E05.4060203@bwh.harvard.edu> Message-ID: Try going through the Volumes module, it has a method CreateLabelVolume which takes a reference greyscale volume as a base (you may need to cast the returns from the Get calls): this->GetApplication()->GetModuleGUIByName("Volumes")->GetLogic()->CreateLabelVolume(...) Nicole On Mon, 9 Mar 2009, James Ross wrote: > Greetings, > > I am working on a module that produces a label map resulting from a > segmentation routine. I would like to strip down the GUI as much as > possible for a specific usage of the module, so I am creating a > vtkMRMLScalarVolumeNode "behind the scenes" (not using a selector), > adding the output segmentation data to it, and then adding the node to > the scene. However, while I see the output node in the MRML tree, the > label map is not rendering. Mousing over different sections of the > input (CT) data, I do see the correct label value indicated, so I > believe that the data has been added to the node, it's just not being > rendered. > > Relevant code: > > void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long > event, void *callData ) > { > ... > this->OutputNode->SetName( "MyModuleNode" ); > this->OutputNode->SetLabelMap( 1 ); > this->OutputNode->SetScene( this->GetMRMLScene() ); > > this->Logic->Apply( ..., this->OutputNode ); // See below > > ... > > this->GetMRMLScene()->AddNode( this->OutputNode ); > ... > } > > > void vtkMyModuleLogic::Apply( ..., vtkMRMLScalarVolumeNode* outputNode ) > { > ... > > outputNode->SetAndObserveDisplayNodeID( inputNode->GetDisplayNodeID() > ); // inputNode is a label map > outputNode->CopyOrientation( inputNode ); > outputNode->SetAndObserveTransformNodeID( > inputNode->GetTransformNodeID() ); > outputNode->SetAndObserveImageData( image ); // image deep copies > filter output > outputNode->SetModifiedSinceRead( 1 ); > > ... > } > > I also tried creating and setting a separate display node for the > OutputNode, but that didn't seem to work either. Any ideas would be > greatly appreciated. > > Regards, > > -James > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > -- Nicole Aucoin nicole at bwh.harvard.edu Surgical Planning Laboratory, Brigham and Women's Hospital Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 http://www.slicer.org From gcsharp at partners.org Tue Mar 10 10:31:20 2009 From: gcsharp at partners.org (Gregory C. Sharp) Date: Tue, 10 Mar 2009 10:31:20 -0400 Subject: [slicer-devel] Python errors during build Message-ID: <49B679B8.7060402@partners.org> Hi, I'm getting some python-related errors during build. This is on Windows 2k. Here is one: error copying "C:/gsharp/build/SLICER~2/Slicer3/../Slicer3-lib/python-build/PCbu ild/_socket.pyd": no such file or directory while executing "file copy -force $::Slicer3_LIB/python-build/PCbuild/_socket.pyd $::Slicer3_LIB /python-build/Lib/_socket.pyd" invoked from within "if { [BuildThis $::PYTHON_TEST_FILE "python"] && !$::USE_SYSTEM_PYTHON && [stri ng tolower $::USE_PYTHON] == "on" } { if { $isWindows } { fi..." (file "./Scripts/genlib.tcl" line 505) Here is another: CMake Error: The following variables are used in this project, but they are set to NOTFOUND. Please set them or make sure they are set and tested correctly in the CMake file s: PYTHON_NUMPY used as include directory in directory C:/gsharp/build/slicer-release/Slicer3 /Base/GUI/Python Thanks, Greg -- Gregory C. Sharp Dept. of Radiation Oncology Massachusetts General Hospital gcsharp at partners.org / 617-724-3866 http://gray.mgh.harvard.edu From fedorov at bwh.harvard.edu Tue Mar 10 10:58:08 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 10 Mar 2009 10:58:08 -0400 Subject: [slicer-devel] Python errors during build In-Reply-To: <49B679B8.7060402@partners.org> References: <49B679B8.7060402@partners.org> Message-ID: <82f536810903100758h7382eb45mc2695210179d8fdd@mail.gmail.com> Gregory, Do you want to use Python-related components on Slicer? If the answer is "no", you can try to disable Python in slicer_variables.tcl to get around the error (look for USE_PYTHON). AF On Tue, Mar 10, 2009 at 10:31 AM, Gregory C. Sharp wrote: > > Hi, > > I'm getting some python-related errors during build. ?This is > on Windows 2k. > > Here is one: > > error copying > "C:/gsharp/build/SLICER~2/Slicer3/../Slicer3-lib/python-build/PCbu > ild/_socket.pyd": no such file or directory > ? ? while executing > "file copy -force $::Slicer3_LIB/python-build/PCbuild/_socket.pyd > $::Slicer3_LIB > /python-build/Lib/_socket.pyd" > ? ? invoked from within > "if { [BuildThis $::PYTHON_TEST_FILE "python"] && !$::USE_SYSTEM_PYTHON > && [stri > ng tolower $::USE_PYTHON] == "on" } { > ? ? if { $isWindows } { > > ? ? ? fi..." > ? ? (file "./Scripts/genlib.tcl" line 505) > > Here is another: > > CMake Error: The following variables are used in this project, but they > are set > to NOTFOUND. > Please set them or make sure they are set and tested correctly in the > CMake file > s: > PYTHON_NUMPY > ? ?used as include directory in directory > C:/gsharp/build/slicer-release/Slicer3 > /Base/GUI/Python > > Thanks, > Greg > -- > Gregory C. Sharp > Dept. of Radiation Oncology > Massachusetts General Hospital > gcsharp at partners.org ?/ ?617-724-3866 > http://gray.mgh.harvard.edu > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From gcsharp at partners.org Tue Mar 10 15:38:28 2009 From: gcsharp at partners.org (Gregory C. Sharp) Date: Tue, 10 Mar 2009 15:38:28 -0400 Subject: [slicer-devel] Python errors during build In-Reply-To: <82f536810903100758h7382eb45mc2695210179d8fdd@mail.gmail.com> References: <49B679B8.7060402@partners.org> <82f536810903100758h7382eb45mc2695210179d8fdd@mail.gmail.com> Message-ID: <49B6C1B4.3020409@partners.org> Andriy Fedorov wrote: > Gregory, > > Do you want to use Python-related components on Slicer? > > If the answer is "no", you can try to disable Python in > slicer_variables.tcl to get around the error (look for USE_PYTHON). Yes, this does work around the errors. Thank you. -Greg -- Gregory C. Sharp Dept. of Radiation Oncology Massachusetts General Hospital gcsharp at partners.org / 617-724-3866 http://gray.mgh.harvard.edu From simon at math.uh.edu Tue Mar 10 16:35:41 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Tue, 10 Mar 2009 15:35:41 -0500 Subject: [slicer-devel] quick question about integration with python or TCL, etc. Message-ID: <200903101535.42102.simon@math.uh.edu> Hi, I'm trying to plan out tighter integration of some of our stuff with Slicer (at the moment, I'm mostly hitting disk and running things elsewhere), and I have a couple of questions for anyone who can enlighten me. Simple one first: is there a way to trigger the module discovery again, not when Slicer starts up. Specifically, if you've made an incompatible change in a scripted module, are you forced to restart Slicer to run it? Secondly, We've been discussing a couple of python based possibilities for above. I've been looking at the various module & plugin possibilities, but I don't have it all sorted yet and it's quite probable that I've missed something. However, it seems to me that the practical (only?) way to have persistent data for anything I'm calling in python would be to stuff it in the MRML tree as a new (or existing if that works) node type. Is my understanding correct? Cheers, Simon Alexander From jross at bwh.harvard.edu Tue Mar 10 17:00:05 2009 From: jross at bwh.harvard.edu (James Ross) Date: Tue, 10 Mar 2009 17:00:05 -0400 Subject: [slicer-devel] Question re. node configuration to display label map data In-Reply-To: References: <49B59E05.4060203@bwh.harvard.edu> Message-ID: <49B6D4D5.6030303@bwh.harvard.edu> Thanks, Nicole. I gave this a try: void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long event, void *callData ) { if ( this->OutputNode == NULL ) { vtkSlicerVolumesLogic* volumesLogic = vtkSlicerVolumesLogic::New(); this->OutputNode = volumesLogic->CreateLabelVolume( this->GetMRMLScene(), vtkMRMLScalarVolumeNode::SafeDownCast( this->GetMRMLScene()->GetNthNode( inputVolumeID ) ), "MyLabelMap" ); } ... } I'm still getting the same behavior, however. I.e., the node is added to the tree (the editor indicates correct settings: label map, origin, etc), and when I mouse over different regions of the input image, I see the correct label value indicated, but adjusting the opacity slider with the label map selected gives no overlay. A potential clue is that when I save the MRML scene and then load it back it, the label map is visible and has the correct behavior. I'm not sure what to make of this. Any ideas? Thanks again for your help. -James Nicole Aucoin wrote: > Try going through the Volumes module, it has a method > CreateLabelVolume which takes a reference greyscale volume as a base > (you may need to cast the returns from the Get calls): > this->GetApplication()->GetModuleGUIByName("Volumes")->GetLogic()->CreateLabelVolume(...) > > > Nicole > > > On Mon, 9 Mar 2009, James Ross wrote: > >> Greetings, >> >> I am working on a module that produces a label map resulting from a >> segmentation routine. I would like to strip down the GUI as much as >> possible for a specific usage of the module, so I am creating a >> vtkMRMLScalarVolumeNode "behind the scenes" (not using a selector), >> adding the output segmentation data to it, and then adding the node to >> the scene. However, while I see the output node in the MRML tree, the >> label map is not rendering. Mousing over different sections of the >> input (CT) data, I do see the correct label value indicated, so I >> believe that the data has been added to the node, it's just not being >> rendered. >> >> Relevant code: >> >> void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long >> event, void *callData ) >> { >> ... >> this->OutputNode->SetName( "MyModuleNode" ); >> this->OutputNode->SetLabelMap( 1 ); >> this->OutputNode->SetScene( this->GetMRMLScene() ); >> >> this->Logic->Apply( ..., this->OutputNode ); // See below >> >> ... >> >> this->GetMRMLScene()->AddNode( this->OutputNode ); >> ... >> } >> >> >> void vtkMyModuleLogic::Apply( ..., vtkMRMLScalarVolumeNode* outputNode ) >> { >> ... >> >> outputNode->SetAndObserveDisplayNodeID( inputNode->GetDisplayNodeID() >> ); // inputNode is a label map >> outputNode->CopyOrientation( inputNode ); >> outputNode->SetAndObserveTransformNodeID( >> inputNode->GetTransformNodeID() ); >> outputNode->SetAndObserveImageData( image ); // image deep copies >> filter output >> outputNode->SetModifiedSinceRead( 1 ); >> >> ... >> } >> >> I also tried creating and setting a separate display node for the >> OutputNode, but that didn't seem to work either. Any ideas would be >> greatly appreciated. >> >> Regards, >> >> -James >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >> > From fedorov at bwh.harvard.edu Tue Mar 10 17:23:41 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 10 Mar 2009 17:23:41 -0400 Subject: [slicer-devel] Question re. node configuration to display label map data In-Reply-To: <49B6D4D5.6030303@bwh.harvard.edu> References: <49B59E05.4060203@bwh.harvard.edu> <49B6D4D5.6030303@bwh.harvard.edu> Message-ID: <82f536810903101423o6dbbf470v1850a3a8be7637b4@mail.gmail.com> James, Does the volume you create appear in the label volume selector? If not, you maybe be missing something as simple as this: slicerApplication->slicerGUI->GetMainSliceGUI("Red")->GetLogic()->GetSliceCompositeNode()->SetLabelVolumeID(this->OutputNode->GetID()); to select the volume. HTH Andriy Fedorov On Tue, Mar 10, 2009 at 5:00 PM, James Ross wrote: > Thanks, Nicole. ?I gave this a try: > > void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long > event, void *callData ) > { > ? ?if ( this->OutputNode == NULL ) > ? ? ?{ > ? ? ?vtkSlicerVolumesLogic* volumesLogic = vtkSlicerVolumesLogic::New(); > > ? ? ?this->OutputNode = volumesLogic->CreateLabelVolume( > this->GetMRMLScene(), > > vtkMRMLScalarVolumeNode::SafeDownCast( this->GetMRMLScene()->GetNthNode( > inputVolumeID ) ), > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?"MyLabelMap" ); > ? ? ?} > ... > > } > > I'm still getting the same behavior, however. ?I.e., the node is added > to the tree (the editor indicates correct settings: label map, origin, > etc), and when I mouse over different regions of the input image, I see > the correct label value indicated, but adjusting the opacity slider with > the label map selected gives no overlay. > > A potential clue is that when I save the MRML scene and then load it > back it, the label map is visible and has the correct behavior. ?I'm not > sure what to make of this. ?Any ideas? > > Thanks again for your help. > > -James > > > > Nicole Aucoin wrote: >> Try going through the Volumes module, it has a method >> CreateLabelVolume which takes a reference greyscale volume as a base >> (you may need to cast the returns from the Get calls): >> this->GetApplication()->GetModuleGUIByName("Volumes")->GetLogic()->CreateLabelVolume(...) >> >> >> Nicole >> >> >> On Mon, 9 Mar 2009, James Ross wrote: >> >>> Greetings, >>> >>> I am working on a module that produces a label map resulting from a >>> segmentation routine. ?I would like to strip down the GUI as much as >>> possible for a specific usage of the module, so I am creating a >>> vtkMRMLScalarVolumeNode "behind the scenes" (not using a selector), >>> adding the output segmentation data to it, and then adding the node to >>> the scene. ?However, while I see the output node in the MRML tree, the >>> label map is not rendering. ?Mousing over different sections of the >>> input (CT) data, I do see the correct label value indicated, so I >>> believe that the data has been added to the node, it's just not being >>> rendered. >>> >>> Relevant code: >>> >>> void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long >>> event, void *callData ) >>> { >>> ? ?... >>> ? ?this->OutputNode->SetName( "MyModuleNode" ); >>> ? ?this->OutputNode->SetLabelMap( 1 ); >>> ? ?this->OutputNode->SetScene( this->GetMRMLScene() ); >>> >>> ? ?this->Logic->Apply( ..., this->OutputNode ); ? // See below >>> >>> ? ?... >>> >>> ? ?this->GetMRMLScene()->AddNode( this->OutputNode ); >>> ? ?... >>> } >>> >>> >>> void vtkMyModuleLogic::Apply( ..., vtkMRMLScalarVolumeNode* outputNode ) >>> { >>> ? ?... >>> >>> ?outputNode->SetAndObserveDisplayNodeID( inputNode->GetDisplayNodeID() >>> ); // inputNode is a label map >>> ?outputNode->CopyOrientation( inputNode ); >>> ?outputNode->SetAndObserveTransformNodeID( >>> inputNode->GetTransformNodeID() ); >>> ?outputNode->SetAndObserveImageData( image ); ?// image deep copies >>> filter output >>> ?outputNode->SetModifiedSinceRead( 1 ); >>> >>> ? ?... >>> } >>> >>> I also tried creating and setting a separate display node for the >>> OutputNode, but that didn't seem to work either. ?Any ideas would be >>> greatly appreciated. >>> >>> Regards, >>> >>> -James >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >>> >> > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From fedorov at bwh.harvard.edu Tue Mar 10 17:25:41 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 10 Mar 2009 17:25:41 -0400 Subject: [slicer-devel] Question re. node configuration to display label map data In-Reply-To: <82f536810903101423o6dbbf470v1850a3a8be7637b4@mail.gmail.com> References: <49B59E05.4060203@bwh.harvard.edu> <49B6D4D5.6030303@bwh.harvard.edu> <82f536810903101423o6dbbf470v1850a3a8be7637b4@mail.gmail.com> Message-ID: <82f536810903101425g30a73035s64969aa8309ceca2@mail.gmail.com> Sorry, I should have sent something more coherent.. here you go: vtkSlicerApplication *application = vtkSlicerApplication::SafeDownCast(this->GetApplication()); vtkSlicerApplicationGUI *applicationGUI = this->GetGUI()->GetApplicationGUI(); applicationGUI->GetMainSliceGUI("Red")->GetLogic()->GetSliceCompositeNode()->SetLabelVolumeID(this->OutputNode->GetID()); On Tue, Mar 10, 2009 at 5:23 PM, Andriy Fedorov wrote: > James, > > Does the volume you create appear in the label volume selector? If > not, you maybe be missing something as simple as this: > > slicerApplication->slicerGUI->GetMainSliceGUI("Red")->GetLogic()->GetSliceCompositeNode()->SetLabelVolumeID(this->OutputNode->GetID()); > > to select the volume. > > HTH > > Andriy Fedorov > > > On Tue, Mar 10, 2009 at 5:00 PM, James Ross wrote: >> Thanks, Nicole. ?I gave this a try: >> >> void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long >> event, void *callData ) >> { >> ? ?if ( this->OutputNode == NULL ) >> ? ? ?{ >> ? ? ?vtkSlicerVolumesLogic* volumesLogic = vtkSlicerVolumesLogic::New(); >> >> ? ? ?this->OutputNode = volumesLogic->CreateLabelVolume( >> this->GetMRMLScene(), >> >> vtkMRMLScalarVolumeNode::SafeDownCast( this->GetMRMLScene()->GetNthNode( >> inputVolumeID ) ), >> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?"MyLabelMap" ); >> ? ? ?} >> ... >> >> } >> >> I'm still getting the same behavior, however. ?I.e., the node is added >> to the tree (the editor indicates correct settings: label map, origin, >> etc), and when I mouse over different regions of the input image, I see >> the correct label value indicated, but adjusting the opacity slider with >> the label map selected gives no overlay. >> >> A potential clue is that when I save the MRML scene and then load it >> back it, the label map is visible and has the correct behavior. ?I'm not >> sure what to make of this. ?Any ideas? >> >> Thanks again for your help. >> >> -James >> >> >> >> Nicole Aucoin wrote: >>> Try going through the Volumes module, it has a method >>> CreateLabelVolume which takes a reference greyscale volume as a base >>> (you may need to cast the returns from the Get calls): >>> this->GetApplication()->GetModuleGUIByName("Volumes")->GetLogic()->CreateLabelVolume(...) >>> >>> >>> Nicole >>> >>> >>> On Mon, 9 Mar 2009, James Ross wrote: >>> >>>> Greetings, >>>> >>>> I am working on a module that produces a label map resulting from a >>>> segmentation routine. ?I would like to strip down the GUI as much as >>>> possible for a specific usage of the module, so I am creating a >>>> vtkMRMLScalarVolumeNode "behind the scenes" (not using a selector), >>>> adding the output segmentation data to it, and then adding the node to >>>> the scene. ?However, while I see the output node in the MRML tree, the >>>> label map is not rendering. ?Mousing over different sections of the >>>> input (CT) data, I do see the correct label value indicated, so I >>>> believe that the data has been added to the node, it's just not being >>>> rendered. >>>> >>>> Relevant code: >>>> >>>> void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long >>>> event, void *callData ) >>>> { >>>> ? ?... >>>> ? ?this->OutputNode->SetName( "MyModuleNode" ); >>>> ? ?this->OutputNode->SetLabelMap( 1 ); >>>> ? ?this->OutputNode->SetScene( this->GetMRMLScene() ); >>>> >>>> ? ?this->Logic->Apply( ..., this->OutputNode ); ? // See below >>>> >>>> ? ?... >>>> >>>> ? ?this->GetMRMLScene()->AddNode( this->OutputNode ); >>>> ? ?... >>>> } >>>> >>>> >>>> void vtkMyModuleLogic::Apply( ..., vtkMRMLScalarVolumeNode* outputNode ) >>>> { >>>> ? ?... >>>> >>>> ?outputNode->SetAndObserveDisplayNodeID( inputNode->GetDisplayNodeID() >>>> ); // inputNode is a label map >>>> ?outputNode->CopyOrientation( inputNode ); >>>> ?outputNode->SetAndObserveTransformNodeID( >>>> inputNode->GetTransformNodeID() ); >>>> ?outputNode->SetAndObserveImageData( image ); ?// image deep copies >>>> filter output >>>> ?outputNode->SetModifiedSinceRead( 1 ); >>>> >>>> ? ?... >>>> } >>>> >>>> I also tried creating and setting a separate display node for the >>>> OutputNode, but that didn't seem to work either. ?Any ideas would be >>>> greatly appreciated. >>>> >>>> Regards, >>>> >>>> -James >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>> subject >>>> >>> >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > From jross at bwh.harvard.edu Tue Mar 10 18:10:14 2009 From: jross at bwh.harvard.edu (James Ross) Date: Tue, 10 Mar 2009 18:10:14 -0400 Subject: [slicer-devel] Question re. node configuration to display label map data In-Reply-To: <82f536810903101425g30a73035s64969aa8309ceca2@mail.gmail.com> References: <49B59E05.4060203@bwh.harvard.edu> <49B6D4D5.6030303@bwh.harvard.edu> <82f536810903101423o6dbbf470v1850a3a8be7637b4@mail.gmail.com> <82f536810903101425g30a73035s64969aa8309ceca2@mail.gmail.com> Message-ID: <49B6E546.5030203@bwh.harvard.edu> Hi Andriy, Thanks -- yes, the volume shows up in the label volume selector. I tried your code as well with no luck. I'm working on a work-around now, but if anyone knows what I'm doing wrong, I'm all ears ;-) -James Andriy Fedorov wrote: > Sorry, I should have sent something more coherent.. here you go: > > vtkSlicerApplication *application = > vtkSlicerApplication::SafeDownCast(this->GetApplication()); > vtkSlicerApplicationGUI *applicationGUI = this->GetGUI()->GetApplicationGUI(); > applicationGUI->GetMainSliceGUI("Red")->GetLogic()->GetSliceCompositeNode()->SetLabelVolumeID(this->OutputNode->GetID()); > > > On Tue, Mar 10, 2009 at 5:23 PM, Andriy Fedorov wrote: > >> James, >> >> Does the volume you create appear in the label volume selector? If >> not, you maybe be missing something as simple as this: >> >> slicerApplication->slicerGUI->GetMainSliceGUI("Red")->GetLogic()->GetSliceCompositeNode()->SetLabelVolumeID(this->OutputNode->GetID()); >> >> to select the volume. >> >> HTH >> >> Andriy Fedorov >> >> >> On Tue, Mar 10, 2009 at 5:00 PM, James Ross wrote: >> >>> Thanks, Nicole. I gave this a try: >>> >>> void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long >>> event, void *callData ) >>> { >>> if ( this->OutputNode == NULL ) >>> { >>> vtkSlicerVolumesLogic* volumesLogic = vtkSlicerVolumesLogic::New(); >>> >>> this->OutputNode = volumesLogic->CreateLabelVolume( >>> this->GetMRMLScene(), >>> >>> vtkMRMLScalarVolumeNode::SafeDownCast( this->GetMRMLScene()->GetNthNode( >>> inputVolumeID ) ), >>> "MyLabelMap" ); >>> } >>> ... >>> >>> } >>> >>> I'm still getting the same behavior, however. I.e., the node is added >>> to the tree (the editor indicates correct settings: label map, origin, >>> etc), and when I mouse over different regions of the input image, I see >>> the correct label value indicated, but adjusting the opacity slider with >>> the label map selected gives no overlay. >>> >>> A potential clue is that when I save the MRML scene and then load it >>> back it, the label map is visible and has the correct behavior. I'm not >>> sure what to make of this. Any ideas? >>> >>> Thanks again for your help. >>> >>> -James >>> >>> >>> >>> Nicole Aucoin wrote: >>> >>>> Try going through the Volumes module, it has a method >>>> CreateLabelVolume which takes a reference greyscale volume as a base >>>> (you may need to cast the returns from the Get calls): >>>> this->GetApplication()->GetModuleGUIByName("Volumes")->GetLogic()->CreateLabelVolume(...) >>>> >>>> >>>> Nicole >>>> >>>> >>>> On Mon, 9 Mar 2009, James Ross wrote: >>>> >>>> >>>>> Greetings, >>>>> >>>>> I am working on a module that produces a label map resulting from a >>>>> segmentation routine. I would like to strip down the GUI as much as >>>>> possible for a specific usage of the module, so I am creating a >>>>> vtkMRMLScalarVolumeNode "behind the scenes" (not using a selector), >>>>> adding the output segmentation data to it, and then adding the node to >>>>> the scene. However, while I see the output node in the MRML tree, the >>>>> label map is not rendering. Mousing over different sections of the >>>>> input (CT) data, I do see the correct label value indicated, so I >>>>> believe that the data has been added to the node, it's just not being >>>>> rendered. >>>>> >>>>> Relevant code: >>>>> >>>>> void vtkMyModuleGUI::ProcessGUIEvents( vtkObject *caller, unsigned long >>>>> event, void *callData ) >>>>> { >>>>> ... >>>>> this->OutputNode->SetName( "MyModuleNode" ); >>>>> this->OutputNode->SetLabelMap( 1 ); >>>>> this->OutputNode->SetScene( this->GetMRMLScene() ); >>>>> >>>>> this->Logic->Apply( ..., this->OutputNode ); // See below >>>>> >>>>> ... >>>>> >>>>> this->GetMRMLScene()->AddNode( this->OutputNode ); >>>>> ... >>>>> } >>>>> >>>>> >>>>> void vtkMyModuleLogic::Apply( ..., vtkMRMLScalarVolumeNode* outputNode ) >>>>> { >>>>> ... >>>>> >>>>> outputNode->SetAndObserveDisplayNodeID( inputNode->GetDisplayNodeID() >>>>> ); // inputNode is a label map >>>>> outputNode->CopyOrientation( inputNode ); >>>>> outputNode->SetAndObserveTransformNodeID( >>>>> inputNode->GetTransformNodeID() ); >>>>> outputNode->SetAndObserveImageData( image ); // image deep copies >>>>> filter output >>>>> outputNode->SetModifiedSinceRead( 1 ); >>>>> >>>>> ... >>>>> } >>>>> >>>>> I also tried creating and setting a separate display node for the >>>>> OutputNode, but that didn't seem to work either. Any ideas would be >>>>> greatly appreciated. >>>>> >>>>> Regards, >>>>> >>>>> -James >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>>> subject >>>>> >>>>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >>> From pieper at bwh.harvard.edu Tue Mar 10 19:50:07 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 10 Mar 2009 19:50:07 -0400 Subject: [slicer-devel] quick question about integration with python or TCL, etc. In-Reply-To: <200903101535.42102.simon@math.uh.edu> References: <200903101535.42102.simon@math.uh.edu> Message-ID: <49B6FCAF.2070800@bwh.harvard.edu> Hi Simon - Simon K Alexander wrote: > Hi, > > I'm trying to plan out tighter integration of some of our stuff with Slicer (at the moment, I'm mostly hitting disk and > running things elsewhere), and I have a couple of questions for anyone who can enlighten me. > > Simple one first: is there a way to trigger the module discovery again, not when Slicer starts up. Specifically, if > you've made an incompatible change in a scripted module, are you forced to restart Slicer to run it? > That's something we've been thinking about but haven't implemented in general. I have done some experiments with this in the Editor module, basically calling EditorTearDownGUI followed by a second call to EditorBuildGUI. This works, but if you have a bug in either of those methods you can still get screwed up. For the GUIs that are generated from the XML it should actually be easier to do (I think there could be something like a ReloadGUI method in the CommandLineModuleGUI class). But for now, I think restarting every time you edit the GUI is the best option. > > Secondly, We've been discussing a couple of python based possibilities for above. I've been looking at the various > module & plugin possibilities, but I don't have it all sorted yet and it's quite probable that I've missed something. > However, it seems to me that the practical (only?) way to have persistent data for anything I'm calling in python would > be to stuff it in the MRML tree as a new (or existing if that works) node type. Is my understanding correct? > Yes, you can create a vtkMRMLScriptedModuleNode in which you can get/set parameters by keyword. These then get stored in the mrml scene (and .mrml file). With regard to the various coding options, there are several 'command line' python examples in Modules/Python and there's a more extensive scripted module example in development here: http://www.nitrc.org/projects/slicervmtklvlst/ Best, Steve > Cheers, > Simon Alexander > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From astakhov at ncmir.ucsd.edu Wed Mar 11 16:33:28 2009 From: astakhov at ncmir.ucsd.edu (vadim astakhov) Date: Wed, 11 Mar 2009 13:33:28 -0700 Subject: [slicer-devel] DTI_image_FiberTrac Message-ID: Is there option to call 3D slicer from a script or java code and get fiber tracs for DTI image. I will appreciate any advice. Vadim Astakhov Center for Research in Biological Systems University of California San Diego From astakhov at ncmir.ucsd.edu Thu Mar 12 14:59:08 2009 From: astakhov at ncmir.ucsd.edu (vadim astakhov) Date: Thu, 12 Mar 2009 11:59:08 -0700 Subject: [slicer-devel] Call_slicer_from_script Message-ID: <51237583-B64A-4F69-9C16-C468BBF29615@ncmir.ucsd.edu> Is there option to call 3D slicer from a script or java code and get fiber tracs for DTI image. I will appreciate any advice. Vadim Astakhov Center for Research in Biological Systems University of California San Diego From simon at math.uh.edu Fri Mar 13 14:53:56 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Fri, 13 Mar 2009 13:53:56 -0500 Subject: [slicer-devel] python command line plugin and fiducial list argument Message-ID: <200903131353.56261.simon@math.uh.edu> Can someone explain to me why the snippet at the end of this email doesn't work? I'm trying to pass a fiducial list as an argument to a python script. I can pass other arguments in similar ways, but when I attempt to pass a "points" argument, I always get the error: Preparing to execute Python ModuleTraceback (most recent call last): File "", line 18, in TypeError: Execute() takes exactly 1 argument (0 given) I've tried all the variations I could think of without any luck. There is a single point example on the wiki but it has a default argument and is for a single point. Any hints? ---- XML = """ DEBUG Modules Points Argument trying to pass a fiducial list as argument 0 Points Test points Fiducial List """ import sys def Execute (points): Slicer = __import__("Slicer") slicer = Slicer.slicer scene = slicer.MRMLScene print '' print 'executed with argument' print point sys.stdout.flush() From simon at math.uh.edu Fri Mar 13 17:32:35 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Fri, 13 Mar 2009 16:32:35 -0500 Subject: [slicer-devel] python command line plugin and fiducial list argument In-Reply-To: <0FE8EDAF-D657-4A56-8976-92C2C23F9E71@sophia.inria.fr> References: <200903131353.56261.simon@math.uh.edu> <0FE8EDAF-D657-4A56-8976-92C2C23F9E71@sophia.inria.fr> Message-ID: <200903131632.36347.simon@math.uh.edu> Hi Demian, That was one of my first thoughts too, but it turns out that's not the problem, something else is going on. The documentation pointed at for the XML section (http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Execution_Model_Documentation:Python) does not suggest that the flag is needed if name is present, but I could have misread it or it may not quite generalize to the Python case. There are differences with how the Module XML is specified in other Modules (non python) anyway. So following the example of the existing Python modules, I first tried with longflag. That doesn't work, it gives the same error. Also indexing is not supported for the point type so that doesn't work. haven't tried just a short flag, I'll try that now. On Friday 13 March 2009 02:56:05 pm you wrote: > Hi Simon, > > you need to provide either an "index" or a "flag" tag to the > object. Index denotes de position in the parameter list > of the execute function starting from 0 (adding 0 > should put your code to work). Adding a "flag" tag tells the name of > a named parameter of the execute function. > > More on this at http://www.slicer.org/slicerWiki/index.php/ > Slicer3:Execution_Model_Documentation > > Stay tuned > > Demian > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > On Mar 13, 2009, at 7:53 PM, Simon K Alexander wrote: > > Can someone explain to me why the snippet at the end of this email > > doesn't work? I'm trying to pass a fiducial list as > > an argument to a python script. > > > > I can pass other arguments in similar ways, but when I attempt to > > pass a "points" argument, I always get the error: > > > > Preparing to execute Python ModuleTraceback (most recent call last): > > File "", line 18, in > > TypeError: Execute() takes exactly 1 argument (0 given) > > > > I've tried all the variations I could think of without any luck. > > There is a single point example on the wiki but it > > has a default argument and is for a single point. > > > > Any hints? > > > > > > ---- > > XML = """ > > > > DEBUG Modules > > Points Argument > > > > trying to pass a fiducial list as argument > > > > 0 > > > > > > > > > > > > > > Points Test > > > > points > > > > Fiducial List > > > > > > > > > > """ > > > > import sys > > > > def Execute (points): > > > > Slicer = __import__("Slicer") > > slicer = Slicer.slicer > > scene = slicer.MRMLScene > > > > print '' > > print 'executed with argument' > > print point > > sys.stdout.flush() > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to slicer-devel- > > request at massmail.spl.harvard.edu with unsubscribe as the subject -- Simon Alexander, Ph.D Department of Mathematics University of Houston 619 PGH phone: 713-743-0220 fax: 713-743-3505 From pieper at bwh.harvard.edu Sat Mar 14 16:04:47 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sat, 14 Mar 2009 16:04:47 -0400 Subject: [slicer-devel] DTI_image_FiberTrac In-Reply-To: References: Message-ID: <49BC0DDF.3050102@bwh.harvard.edu> Hi Vadim - To generate tracts from a script, you could either use the command line executable for seeding that comes with slicer: http://viewvc.slicer.org/viewcvs.cgi/trunk/Modules/Tractography/Seeding/?rev=8877 The command line would be something like: ./Slicer3 --launch lib/Slicer3/Plugins/Seeding or you could call the underlying classes directly from a tcl or python inside slicer. Also, if you have fiber bundles visible in the slicer scene you can access them via tcl or python. See this code from Demian, for example, which accesses the underlying point arrays and manipulates them with numpy: http://www.slicer.org/slicerWiki/index.php/Slicer3:Python:DemianExamples#K-Medoids_Fiber_Clustering Best, Steve vadim astakhov wrote: > Is there option to call 3D slicer from a script or java code and get > fiber tracs for DTI image. > > I will appreciate any advice. > > Vadim Astakhov > > Center for Research in Biological Systems > University of California San Diego > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Sat Mar 14 18:59:49 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sat, 14 Mar 2009 18:59:49 -0400 Subject: [slicer-devel] python command line plugin and fiducial list argument In-Reply-To: <200903131632.36347.simon@math.uh.edu> References: <200903131353.56261.simon@math.uh.edu> <0FE8EDAF-D657-4A56-8976-92C2C23F9E71@sophia.inria.fr> <200903131632.36347.simon@math.uh.edu> Message-ID: <49BC36E5.20401@bwh.harvard.edu> Hi Simon - I was able to get this working by adding the long flag and then a corresponding argument to the Execute function. Does this do what you had in mind? -Steve XML = """ DEBUG Modules Points Argument trying to pass a fiducial list as argument 0 Points Test points points Fiducial List """ import sys def Execute (points=""): Slicer = __import__("Slicer") slicer = Slicer.slicer scene = slicer.MRMLScene print '' print 'executed with argument' print point sys.stdout.flush() Simon K Alexander wrote: > Hi Demian, > > That was one of my first thoughts too, but it turns out that's not the problem, something else is going on. > > The documentation pointed at for the XML section > (http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Execution_Model_Documentation:Python) does not suggest that > the flag is needed if name is present, but I could have misread it or it may not quite generalize to the Python case. > There are differences with how the Module XML is specified in other Modules (non python) anyway. > > So following the example of the existing Python modules, I first tried with longflag. That doesn't work, it gives the > same error. Also indexing is not supported for the point type so that doesn't work. haven't tried just a short flag, > I'll try that now. > > > On Friday 13 March 2009 02:56:05 pm you wrote: >> Hi Simon, >> >> you need to provide either an "index" or a "flag" tag to the >> object. Index denotes de position in the parameter list >> of the execute function starting from 0 (adding 0 >> should put your code to work). Adding a "flag" tag tells the name of >> a named parameter of the execute function. >> >> More on this at http://www.slicer.org/slicerWiki/index.php/ >> Slicer3:Execution_Model_Documentation >> >> Stay tuned >> >> Demian >> >> -- >> Demian Wassermann >> demian.wassermann at sophia.inria.fr >> PhD Student >> Odyssee Research Project >> INRIA Sophia-Antipolis >> 2004 route des lucioles - FR-06902 >> >> On Mar 13, 2009, at 7:53 PM, Simon K Alexander wrote: >>> Can someone explain to me why the snippet at the end of this email >>> doesn't work? I'm trying to pass a fiducial list as >>> an argument to a python script. >>> >>> I can pass other arguments in similar ways, but when I attempt to >>> pass a "points" argument, I always get the error: >>> >>> Preparing to execute Python ModuleTraceback (most recent call last): >>> File "", line 18, in >>> TypeError: Execute() takes exactly 1 argument (0 given) >>> >>> I've tried all the variations I could think of without any luck. >>> There is a single point example on the wiki but it >>> has a default argument and is for a single point. >>> >>> Any hints? >>> >>> >>> ---- >>> XML = """ >>> >>> DEBUG Modules >>> Points Argument >>> >>> trying to pass a fiducial list as argument >>> >>> 0 >>> >>> >>> >>> >>> >>> >>> Points Test >>> >>> points >>> >>> Fiducial List >>> >>> >>> >>> >>> """ >>> >>> import sys >>> >>> def Execute (points): >>> >>> Slicer = __import__("Slicer") >>> slicer = Slicer.slicer >>> scene = slicer.MRMLScene >>> >>> print '' >>> print 'executed with argument' >>> print point >>> sys.stdout.flush() >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel- >>> request at massmail.spl.harvard.edu with unsubscribe as the subject > > > From simon at math.uh.edu Sun Mar 15 13:42:52 2009 From: simon at math.uh.edu (Simon Alexander) Date: Sun, 15 Mar 2009 12:42:52 -0500 Subject: [slicer-devel] build issue with local python Message-ID: Hi, I ran into a strange build issue (OS X) when I tried a build using the system python. Setting USE_SYSTEM_PYTHON to "true", and USE_NUMPY "false" (not sure if the latter makes a difference), I then did a clean build. The resulting binary will attempt to run strangely enough has errors with the TCL setup, not the python! > simon-2:Slicer3 simon$ DISPLAY=:0 ../Slicer3-build/Slicer3 > Tcl_Init error: Can't find a usable init.tcl in the following > directories: > /Users/simon/build/Slicer3-build/lib/TclTk/lib/tcl8.4 > @TCL_IN_FRAMEWORK@ /Users/simon/build/Slicer3-build/lib/TclTk/lib/ > tcl8.4 /Users/simon/build/Slicer3-build/lib/tcl8.4 /Users/simon/ > build/lib/tcl8.4 /Users/simon/build/Slicer3-build/library /Users/ > simon/build/library /Users/simon/build/tcl8.4.7/library /Users/simon/ > tcl8.4.7/library @TCL_IN_FRAMEWORK@ Any ideas? From simon at math.uh.edu Sun Mar 15 14:02:56 2009 From: simon at math.uh.edu (Simon Alexander) Date: Sun, 15 Mar 2009 13:02:56 -0500 Subject: [slicer-devel] python command line plugin and fiducial list argument In-Reply-To: <49BD3B4A.5030003@bwh.harvard.edu> References: <200903131353.56261.simon@math.uh.edu> <0FE8EDAF-D657-4A56-8976-92C2C23F9E71@sophia.inria.fr> <200903131632.36347.simon@math.uh.edu> <49BC36E5.20401@bwh.harvard.edu> <85DD7BD7-32F1-4B21-B7B9-C3C4D73EF2E5@math.uh.edu> <49BD3B4A.5030003@bwh.harvard.edu> Message-ID: <4BD0C643-9BB0-4596-B191-9E6423CC47A3@math.uh.edu> Hi Steve, Thanks for looking into this. Your success made me dig a little deeper, and I think I've figured out what's going on. First, the only way I can get anything passed is the longflag method but when I first tried this (as noted previously) it didn't work. It turns out (and I think I'd classify this as a bug) that if you specify a list where none of the fiducials are selected, rather than passing an empty list it doesn't pass anything at all. So you get the default argument if it exists, or a failure if there is no default. When I was debugging this to save time I was just firing up a new Slicer and clicking a few points to create a new Fiducial list, but without a volume loaded (unlike the case when it is selected) the points are created in an unselected state. So my whole list was unselected. Should we file this as a bug? It's a weird edge case, but will break any script without a default points specified. cheers, Simon On a related note, is there any way to get the ID (or whatever) of the actual Fiducial list passed to a filter like this, rather than a list of the points? On Mar 15, 2009, at 12:30 PM, Steve Pieper wrote: > Hi Simon - > > Looks like I pasted the wrong version that (with the print of > 'point' rather than 'points' - oops!). > > With the fix I do get the correct output when I pass in a fiducial > list:: > >> executed with argument >> [[-56.506799999999998, 17.1233, 0.0], [12.5571, >> -54.223700000000001, 0.0]] > > This should be the right code this time! ;) > > -Steve From pieper at bwh.harvard.edu Sun Mar 15 14:10:26 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sun, 15 Mar 2009 14:10:26 -0400 Subject: [slicer-devel] build issue with local python In-Reply-To: References: Message-ID: <49BD4492.2020606@bwh.harvard.edu> Hi Simon - I haven't tried the USE_SYSTEM_PYTHON myself, but something in the startup process seems to be overwriting the TCL_LIBRARY environment variable. If you run this: ./Slicer3 --launch env | grep TCL_LIBRARY you should get something like this that tells where slicer expects to find the init.tcl file: TCL_LIBRARY=/Users/pieper/slicer3/latest/Slicer3-lib/tcl-build/lib/tcl8.4 Python does rely on the tcl libraries internally for some things so there could be problems when the versions mix and match, particularly with the mac frameworks and related issues. That @TCL_IN_FRAMEWORK@ symbol is particularly suspicious (almost like something didn't get built correctly...). -Steve Simon Alexander wrote: > Hi, > > I ran into a strange build issue (OS X) when I tried a build using the > system python. > > Setting USE_SYSTEM_PYTHON to "true", and USE_NUMPY "false" (not sure > if the latter makes a difference), I then did a clean build. > > The resulting binary will attempt to run strangely enough has errors > with the TCL setup, not the python! > >> simon-2:Slicer3 simon$ DISPLAY=:0 ../Slicer3-build/Slicer3 >> Tcl_Init error: Can't find a usable init.tcl in the following >> directories: >> /Users/simon/build/Slicer3-build/lib/TclTk/lib/tcl8.4 >> @TCL_IN_FRAMEWORK@ /Users/simon/build/Slicer3-build/lib/TclTk/lib/ >> tcl8.4 /Users/simon/build/Slicer3-build/lib/tcl8.4 /Users/simon/ >> build/lib/tcl8.4 /Users/simon/build/Slicer3-build/library /Users/ >> simon/build/library /Users/simon/build/tcl8.4.7/library /Users/simon/ >> tcl8.4.7/library @TCL_IN_FRAMEWORK@ > > > Any ideas? > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Sun Mar 15 14:25:59 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sun, 15 Mar 2009 14:25:59 -0400 Subject: [slicer-devel] python command line plugin and fiducial list argument In-Reply-To: <4BD0C643-9BB0-4596-B191-9E6423CC47A3@math.uh.edu> References: <200903131353.56261.simon@math.uh.edu> <0FE8EDAF-D657-4A56-8976-92C2C23F9E71@sophia.inria.fr> <200903131632.36347.simon@math.uh.edu> <49BC36E5.20401@bwh.harvard.edu> <85DD7BD7-32F1-4B21-B7B9-C3C4D73EF2E5@math.uh.edu> <49BD3B4A.5030003@bwh.harvard.edu> <4BD0C643-9BB0-4596-B191-9E6423CC47A3@math.uh.edu> Message-ID: <49BD4837.7070304@bwh.harvard.edu> Hi Simon - Yes, I agree it should be sending an empty list rather than ignoring the argument. If you could dash off a quick bug report that will keep it in the queue. For now specifying the default value is probably an okay workaround. Right now I think only the list of points is available, but being able to get a the fiducial list ID into the python module would be another nice feature to add. Best, Steve Simon Alexander wrote: > Hi Steve, > > Thanks for looking into this. Your success made me dig a little deeper, > and I think I've figured out what's going on. First, the only way I > can get anything passed is the longflag method but when I first tried > this (as noted previously) it didn't work. > > It turns out (and I think I'd classify this as a bug) that if you > specify a list where none of the fiducials are selected, rather than > passing an empty list it doesn't pass anything at all. So you get the > default argument if it exists, or a failure if there is no default. > > When I was debugging this to save time I was just firing up a new Slicer > and clicking a few points to create a new Fiducial list, but without a > volume loaded (unlike the case when it is selected) the points are > created in an unselected state. So my whole list was unselected. > > Should we file this as a bug? It's a weird edge case, but will break > any script without a default points specified. > > > cheers, > Simon > > > > On a related note, is there any way to get the ID (or whatever) of the > actual Fiducial list passed to a filter like this, rather than a list of > the points? > > On Mar 15, 2009, at 12:30 PM, Steve Pieper wrote: > >> Hi Simon - >> >> Looks like I pasted the wrong version that (with the print of 'point' >> rather than 'points' - oops!). >> >> With the fix I do get the correct output when I pass in a fiducial list:: >> >>> executed with argument >>> [[-56.506799999999998, 17.1233, 0.0], [12.5571, -54.223700000000001, >>> 0.0]] >> >> This should be the right code this time! ;) >> >> -Steve From pieper at bwh.harvard.edu Sun Mar 15 16:16:36 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sun, 15 Mar 2009 16:16:36 -0400 Subject: [slicer-devel] system python packages - Re: build issue with local python In-Reply-To: <49BD4492.2020606@bwh.harvard.edu> References: <49BD4492.2020606@bwh.harvard.edu> Message-ID: <49BD6224.7050004@bwh.harvard.edu> By the way, you can also get access to scipy and matplotlib just by setting the path and using the binaries downloaded from scipy.org (on the mac anyway - I haven't tied other platforms). This is still 'iffy' but should work for testing since scipy and slicer both build against python2.5 and tcl8.4. Longer term a goal is to get all these packages to build as part of slicer. For example, the code below works for me. -Steve import sys d = '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] for p in packages: sys.path.append(d+p) import matplotlib matplotlib.use ( 'Agg' ) from pylab import * t1 = arange(0.0, 5.0, 0.1) t2 = arange(0.0, 5.0, 0.02) t3 = arange(0.0, 2.0, 0.01) subplot(211) plot(t1, cos(2*pi*t1)*exp(-t1), 'bo', t2, cos(2*pi*t2)*exp(-t2), 'k') grid(True) title('A tale of 2 subplots') ylabel('Damped') subplot(212) plot(t3, cos(2*pi*t3), 'r--') grid(True) xlabel('time (s)') ylabel('Undamped') savefig ( 'MatplotlibExample.png' ) import Slicer r = Slicer.slicer.vtkPNGReader() v = Slicer.slicer.vtkImageViewer() r.SetFileName( 'MatplotlibExample.png' ) v.SetColorWindow(255) v.SetColorLevel(128) v.SetInput(r.GetOutput()) v.Render() Steve Pieper wrote: > Hi Simon - > > I haven't tried the USE_SYSTEM_PYTHON myself, but something in the > startup process seems to be overwriting the TCL_LIBRARY environment > variable. > > If you run this: > ./Slicer3 --launch env | grep TCL_LIBRARY > > you should get something like this that tells where slicer expects to > find the init.tcl file: > > TCL_LIBRARY=/Users/pieper/slicer3/latest/Slicer3-lib/tcl-build/lib/tcl8.4 > > Python does rely on the tcl libraries internally for some things so > there could be problems when the versions mix and match, particularly > with the mac frameworks and related issues. That @TCL_IN_FRAMEWORK@ > symbol is particularly suspicious (almost like something didn't get > built correctly...). > > -Steve > > Simon Alexander wrote: >> Hi, >> >> I ran into a strange build issue (OS X) when I tried a build using the >> system python. >> >> Setting USE_SYSTEM_PYTHON to "true", and USE_NUMPY "false" (not sure >> if the latter makes a difference), I then did a clean build. >> >> The resulting binary will attempt to run strangely enough has errors >> with the TCL setup, not the python! >> >>> simon-2:Slicer3 simon$ DISPLAY=:0 ../Slicer3-build/Slicer3 >>> Tcl_Init error: Can't find a usable init.tcl in the following >>> directories: >>> /Users/simon/build/Slicer3-build/lib/TclTk/lib/tcl8.4 >>> @TCL_IN_FRAMEWORK@ /Users/simon/build/Slicer3-build/lib/TclTk/lib/ >>> tcl8.4 /Users/simon/build/Slicer3-build/lib/tcl8.4 /Users/simon/ >>> build/lib/tcl8.4 /Users/simon/build/Slicer3-build/library /Users/ >>> simon/build/library /Users/simon/build/tcl8.4.7/library /Users/simon/ >>> tcl8.4.7/library @TCL_IN_FRAMEWORK@ >> >> Any ideas? >> >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From wschramm at bwh.harvard.edu Tue Mar 17 16:50:33 2009 From: wschramm at bwh.harvard.edu (Wolfgang Schramm) Date: Tue, 17 Mar 2009 16:50:33 -0400 Subject: [slicer-devel] Create LabelMap from Model? Message-ID: <49C00D19.9050300@bwh.harvard.edu> Hello, I'm currently trying to create a labelmap from a 3D Model (simple sphere and cylinder). Is there a way to do it in the Slicer UI, or can anybody point me to the correct vtk/slicer classes to use? Thank you Wolfgang From inorton at partners.org Tue Mar 17 17:45:16 2009 From: inorton at partners.org (Isaiah Norton) Date: Tue, 17 Mar 2009 17:45:16 -0400 Subject: [slicer-devel] Create LabelMap from Model? In-Reply-To: <49C00D19.9050300@bwh.harvard.edu> References: <49C00D19.9050300@bwh.harvard.edu> Message-ID: <1237326316.5051.26.camel@thanks.bwh.harvard.edu> Hi Wolfgang, This module might help: Model Generation->Model_Into_Label_Volume. This colors voxels by polydata vertices so will give you the outline of the model (not the middle) The above was designed specifically to label tract-volume intersections so YMMV with other models (worked for a basic sphereish tumor model). It might be inefficient for a very dense model b/c it searches every vertex. I think the vtk function to do this is vtkImplicitFunction. -I On Tue, 2009-03-17 at 16:50 -0400, Wolfgang Schramm wrote: > Hello, > > I'm currently trying to create a labelmap from a 3D Model (simple sphere > and cylinder). Is there a way to do it in the Slicer UI, or can anybody > point me to the correct vtk/slicer classes to use? > > Thank you > Wolfgang > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > > > The information in this e-mail is intended only for the person to whom it is > addressed. If you believe this e-mail was sent to you in error and the e-mail > contains patient information, please contact the Partners Compliance HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent to you in error > but does not contain patient information, please contact the sender and properly > dispose of the e-mail. > From daniel.blezek at gmail.com Tue Mar 17 21:43:28 2009 From: daniel.blezek at gmail.com (Daniel Blezek) Date: Tue, 17 Mar 2009 20:43:28 -0500 Subject: [slicer-devel] Commit problems Message-ID: <1CC66820-5B12-4DD1-BC40-1E8DC3AC3379@gmail.com> Hi, I've been trying to checkin code to support OpenSSL in curl, however, I get an error message: Error: Commit failed (details follow): Error: Server sent unexpected return value (500 Internal Server Error) in response to Error: MKACTIVITY request for '/svn/Slicer3-lib-mirrors/!svn/act/ 282927e4-0262-ec40-ba31-139bd9fca5b2' Is this a problem on my end? The code goes in: http://www.na-mic.org/svn/Slicer3-lib-mirrors/trunk/cmcurl Thanks, -dan -- Daniel Blezek daniel.blezek at gmail.com From vrnova at gmail.com Wed Mar 18 09:03:21 2009 From: vrnova at gmail.com (Yanling Liu) Date: Wed, 18 Mar 2009 09:03:21 -0400 Subject: [slicer-devel] AddObserver and SetCommand Message-ID: <24ceacd90903180603s44a46c0eqdb71e1cdc1dd0aec@mail.gmail.com> Hello, is there a guideline to help to decide when to use AddObserver and SetCommand for KWWidgets? I have some widgets respond only to AddObserver and if I try to use SetCommand the command function will never be called. Here is an example: //this one works, the VolumeRenderingCallbackCommand catches the event this->SC_ExpectedFPS->AddObserver(vtkKWScale::ScaleValueChangedEvent, (vtkCommand *) this->VolumeRenderingCallbackCommand); //this one doesn't work, the ProcessExpectedFPS never got called this->SC_ExpectedFPS->SetCommand(this, "ProcessExpectedFPS"); Is there a hint? Thanks a lot! Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090318/20215558/attachment-0001.html From pieper at bwh.harvard.edu Wed Mar 18 12:16:00 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 18 Mar 2009 12:16:00 -0400 Subject: [slicer-devel] Commit problems In-Reply-To: <1CC66820-5B12-4DD1-BC40-1E8DC3AC3379@gmail.com> References: <1CC66820-5B12-4DD1-BC40-1E8DC3AC3379@gmail.com> Message-ID: <49C11E40.1010706@bwh.harvard.edu> Hi Dan - I just checked and I'm able to update and commit in that directory, so maybe it's something about your account. I'll cc Zack on this and he can probably troubleshoot it if you're still having problems. -Steve Daniel Blezek wrote: > Hi, > > I've been trying to checkin code to support OpenSSL in curl, > however, I get an error message: > > Error: Commit failed (details follow): > Error: Server sent unexpected return value (500 Internal Server Error) > in response to > Error: MKACTIVITY request for '/svn/Slicer3-lib-mirrors/!svn/act/ > 282927e4-0262-ec40-ba31-139bd9fca5b2' > > Is this a problem on my end? The code goes in: http://www.na-mic.org/svn/Slicer3-lib-mirrors/trunk/cmcurl > > Thanks, > -dan > > -- > Daniel Blezek > daniel.blezek at gmail.com > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > > > The information in this e-mail is intended only for the person to whom it is > addressed. If you believe this e-mail was sent to you in error and the e-mail > contains patient information, please contact the Partners Compliance HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent to you in error > but does not contain patient information, please contact the sender and properly > dispose of the e-mail. > From wschramm at bwh.harvard.edu Wed Mar 18 12:33:10 2009 From: wschramm at bwh.harvard.edu (Wolfgang Schramm) Date: Wed, 18 Mar 2009 12:33:10 -0400 Subject: [slicer-devel] =?utf-8?q?Create_LabelMap_from_Model=3F?= In-Reply-To: <1237326316.5051.26.camel@thanks.bwh.harvard.edu> References: <1237326316.5051.26.camel@thanks.bwh.harvard.edu> Message-ID: <94f5b98593c12c2f85e0df4b3a5a907d@bwh.harvard.edu> Hi Isaiah, thank you for your input. ModelIntoLabelVolume looks very promising, but it doesn't work 100% for me. It works fine with the sphere model, but it shows no result for my cylinder. I assume my method of creating the cylinder is the cause for this: 1) Create two points (I use fiducials for that purpose) 2) Create a lineSource between these points 3) Create the cylinder by feeding the linesource through a tube filter. Do you have any idea on how to circumvent this problem? Thank you Wolfgang On Tue, 17 Mar 2009 17:45:16 -0400, Isaiah Norton wrote: > Hi Wolfgang, > > This module might help: Model Generation->Model_Into_Label_Volume. This > colors voxels by polydata vertices so will give you the outline of the > model (not the middle) > > The above was designed specifically to label tract-volume intersections > so YMMV with other models (worked for a basic sphereish tumor model). It > might be inefficient for a very dense model b/c it searches every > vertex. > > I think the vtk function to do this is vtkImplicitFunction. > > -I > > > On Tue, 2009-03-17 at 16:50 -0400, Wolfgang Schramm wrote: >> Hello, >> >> I'm currently trying to create a labelmap from a 3D Model (simple sphere >> and cylinder). Is there a way to do it in the Slicer UI, or can anybody >> point me to the correct vtk/slicer classes to use? >> >> Thank you >> Wolfgang >> >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject >> >> >> The information in this e-mail is intended only for the person to whom > it is >> addressed. If you believe this e-mail was sent to you in error and the > e-mail >> contains patient information, please contact the Partners Compliance > HelpLine at >> http://www.partners.org/complianceline . If the e-mail was sent to you > in error >> but does not contain patient information, please contact the sender and > properly >> dispose of the e-mail. >> > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject From pieper at bwh.harvard.edu Wed Mar 18 12:44:42 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 18 Mar 2009 12:44:42 -0400 Subject: [slicer-devel] AddObserver and SetCommand In-Reply-To: <24ceacd90903180603s44a46c0eqdb71e1cdc1dd0aec@mail.gmail.com> References: <24ceacd90903180603s44a46c0eqdb71e1cdc1dd0aec@mail.gmail.com> Message-ID: <49C124FA.3050008@bwh.harvard.edu> Hi Yanling - I prefer to use the even observation model for most things since it is consistent with the events from other vtk clases (mrml etc). I believe most things in KWWidgets issue events and call commands (see example below) but some things may only do one or the other. -Steve //---------------------------------------------------------------------------- void vtkKWScale::InvokeCommand(double value) { this->InvokeScaleCommand(this->Command, value); this->InvokeEvent(vtkKWScale::ScaleValueChangingEvent, &value); } Yanling Liu wrote: > Hello, is there a guideline to help to decide when to use AddObserver > and SetCommand for KWWidgets? I have some widgets respond only to > AddObserver and if I try to use SetCommand the command function will > never be called. Here is an example: > > //this one works, the VolumeRenderingCallbackCommand catches the event > this->SC_ExpectedFPS->AddObserver(vtkKWScale::ScaleValueChangedEvent, > (vtkCommand *) this->VolumeRenderingCallbackCommand); > > //this one doesn't work, the ProcessExpectedFPS never got called > this->SC_ExpectedFPS->SetCommand(this, "ProcessExpectedFPS"); > > Is there a hint? > > Thanks a lot! > Yanling > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > > > The information in this e-mail is intended only for the person to whom it is > addressed. If you believe this e-mail was sent to you in error and the e-mail > contains patient information, please contact the Partners Compliance HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent to you in error > but does not contain patient information, please contact the sender and properly > dispose of the e-mail. From cvachet at email.unc.edu Thu Mar 19 11:25:20 2009 From: cvachet at email.unc.edu (cvachet at email.unc.edu) Date: Thu, 19 Mar 2009 11:25:20 -0400 Subject: [slicer-devel] RegisterImages availability Message-ID: <20090319112520.t50ccom1mo48c80g@webmail5.isis.unc.edu> Dear all, When we compile Slicer3 from the source code, RegisterImages is compiled by default. However RegisterImages doesn't belong to Slicer3 nigthly builds. Is there a specific reason? In our regional cortical thickness analysis pipeline, ARCTIC (Slicer3 external module) runs several Slicer modules: RegisterImages, ResampleVolume2, ModelMaker. As we mostly use nightly builds versions of Slicer, we would be happy to have RegisterImages available. Thanks, Best regards, Clement Vachet. From kikinis at bwh.harvard.edu Thu Mar 19 12:30:13 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Thu, 19 Mar 2009 12:30:13 -0400 Subject: [slicer-devel] RegisterImages availability In-Reply-To: <20090319112520.t50ccom1mo48c80g@webmail5.isis.unc.edu> References: <20090319112520.t50ccom1mo48c80g@webmail5.isis.unc.edu> Message-ID: <49C27315.4040809@bwh.harvard.edu> Clement, Register images is included in the nightly build at least on the MAC. Ron cvachet at email.unc.edu wrote: > Dear all, > > When we compile Slicer3 from the source code, RegisterImages is > compiled by default. However RegisterImages doesn't belong to Slicer3 > nigthly builds. Is there a specific reason? In our regional cortical > thickness analysis pipeline, ARCTIC (Slicer3 external module) runs > several Slicer modules: RegisterImages, ResampleVolume2, ModelMaker. As > we mostly use nightly builds versions of Slicer, we would be happy to > have RegisterImages available. > > Thanks, > Best regards, > Clement Vachet. > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > > > The information in this e-mail is intended only for the person to whom it is > addressed. If you believe this e-mail was sent to you in error and the e-mail > contains patient information, please contact the Partners Compliance HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent to you in error > but does not contain patient information, please contact the sender and properly > dispose of the e-mail. > -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From hayes at bwh.harvard.edu Thu Mar 19 12:47:17 2009 From: hayes at bwh.harvard.edu (Kathryn Hayes) Date: Thu, 19 Mar 2009 12:47:17 -0400 (EDT) Subject: [slicer-devel] RegisterImages availability In-Reply-To: <49C27315.4040809@bwh.harvard.edu> References: <20090319112520.t50ccom1mo48c80g@webmail5.isis.unc.edu> <49C27315.4040809@bwh.harvard.edu> Message-ID: I just untarred last night's Linux 64 build, and it looks like the tarball is corrupted right in the middle of the RegisterImages shared object: tar: Slicer3-3.3-alpha-2009-03-18-linux-x86_64/lib/Slicer3/Plugins/libRegisterImagesModule.so: Wrote only 3584 of 10240 bytes tar: Skipping to next header tar: Archive contains obsolescent base-64 headers tar: Error exit delayed from previous errors I'm guessing this is the problem. Let me rebuild on Linux and see if that fixes it. Katie On Thu, 19 Mar 2009, Ron Kikinis wrote: > Clement, > Register images is included in the nightly build at least on the MAC. > Ron > > cvachet at email.unc.edu wrote: >> Dear all, >> >> When we compile Slicer3 from the source code, RegisterImages is >> compiled by default. However RegisterImages doesn't belong to Slicer3 >> nigthly builds. Is there a specific reason? In our regional cortical >> thickness analysis pipeline, ARCTIC (Slicer3 external module) runs >> several Slicer modules: RegisterImages, ResampleVolume2, ModelMaker. As >> we mostly use nightly builds versions of Slicer, we would be happy to >> have RegisterImages available. >> >> Thanks, >> Best regards, >> Clement Vachet. >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> >> >> The information in this e-mail is intended only for the person to whom it is >> addressed. If you believe this e-mail was sent to you in error and the e-mail >> contains patient information, please contact the Partners Compliance HelpLine at >> http://www.partners.org/complianceline . If the e-mail was sent to you in error >> but does not contain patient information, please contact the sender and properly >> dispose of the e-mail. >> > > From hayes at bwh.harvard.edu Thu Mar 19 13:06:47 2009 From: hayes at bwh.harvard.edu (Kathryn Hayes) Date: Thu, 19 Mar 2009 13:06:47 -0400 (EDT) Subject: [slicer-devel] RegisterImages availability In-Reply-To: References: <20090319112520.t50ccom1mo48c80g@webmail5.isis.unc.edu> <49C27315.4040809@bwh.harvard.edu> Message-ID: Can you try this one and see if you're still having trouble? It works for me. http://www.na-mic.org/Slicer/Download/Nightly/Slicer3-3.3-alpha-2009-03-19-linux-x86_64.tar.gz Thanks, Katie On Thu, 19 Mar 2009, Kathryn Hayes wrote: > I just untarred last night's Linux 64 build, and it looks like the tarball > is corrupted right in the middle of the RegisterImages shared object: > > tar: > Slicer3-3.3-alpha-2009-03-18-linux-x86_64/lib/Slicer3/Plugins/libRegisterImagesModule.so: > Wrote only 3584 of 10240 bytes > tar: Skipping to next header > tar: Archive contains obsolescent base-64 headers > tar: Error exit delayed from previous errors > > I'm guessing this is the problem. Let me rebuild on Linux and see if that > fixes it. > > Katie > > On Thu, 19 Mar 2009, Ron Kikinis wrote: > >> Clement, >> Register images is included in the nightly build at least on the MAC. >> Ron >> >> cvachet at email.unc.edu wrote: >>> Dear all, >>> >>> When we compile Slicer3 from the source code, RegisterImages is >>> compiled by default. However RegisterImages doesn't belong to Slicer3 >>> nigthly builds. Is there a specific reason? In our regional cortical >>> thickness analysis pipeline, ARCTIC (Slicer3 external module) runs >>> several Slicer modules: RegisterImages, ResampleVolume2, ModelMaker. As >>> we mostly use nightly builds versions of Slicer, we would be happy to >>> have RegisterImages available. >>> >>> Thanks, >>> Best regards, >>> Clement Vachet. >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >>> >>> The information in this e-mail is intended only for the person to whom it is >>> addressed. If you believe this e-mail was sent to you in error and the e-mail >>> contains patient information, please contact the Partners Compliance HelpLine at >>> http://www.partners.org/complianceline . If the e-mail was sent to you in error >>> but does not contain patient information, please contact the sender and properly >>> dispose of the e-mail. >>> >> >> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From pieper at bwh.harvard.edu Thu Mar 19 15:46:56 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 19 Mar 2009 15:46:56 -0400 Subject: [slicer-devel] menu reorganization & 3.4 release planning Message-ID: <49C2A130.3040505@bwh.harvard.edu> Hi Everyone - As you know, several of us are hard at work cleaning things up for a 3.4 release of slicer. There will be at least another few weeks of things to clear up before we create the release branch. Here are a few items people should be aware of: ** tagging bugs ** Rather than pick an arbitrary release date, we are attaching the tag "3.4 Targeted fix" to any bug we plan to fix in this version. We'll know the release is ready when anything tagged with this flag is closed. Developers: please go through your active bugs and apply this tag. Users: if you have issues that you think should be resolved, please contact the developer and encourage them to tag your bugs for the release. ** reorganized modules ** The main modules menu has been reorganized to match the layout in the documentation: http://www.slicer.org/slicerWiki/index.php/Documentation-3.4 This required some changes to module categories and (sometimes) module names themselves. You can see the specific changes in checkin below: http://viewvc.slicer.org/viewcvs.cgi?rev=8915&view=rev This new layout will start appearing in tomorrows nightly builds. Best, Steve p.s. many thanks to everyone for all the hard work testing, bug reporting, and bug fixing! From cvachet at email.unc.edu Thu Mar 19 16:00:14 2009 From: cvachet at email.unc.edu (cvachet at email.unc.edu) Date: Thu, 19 Mar 2009 16:00:14 -0400 Subject: [slicer-devel] RegisterImages availability In-Reply-To: References: <20090319112520.t50ccom1mo48c80g@webmail5.isis.unc.edu> <49C27315.4040809@bwh.harvard.edu> Message-ID: <20090319160014.ksxxhddiosksw848@webmail5.isis.unc.edu> Thanks Katie and Ron for the info! RegisterImages is indeed available in the latest 64bits nightly build (03/19). Best regards, Clement. Quoting Kathryn Hayes : > Can you try this one and see if you're still having trouble? It > works for me. > > http://www.na-mic.org/Slicer/Download/Nightly/Slicer3-3.3-alpha-2009-03-19-linux-x86_64.tar.gz > > Thanks, > > Katie > > On Thu, 19 Mar 2009, Kathryn Hayes wrote: > >> I just untarred last night's Linux 64 build, and it looks like the tarball >> is corrupted right in the middle of the RegisterImages shared object: >> >> tar: >> Slicer3-3.3-alpha-2009-03-18-linux-x86_64/lib/Slicer3/Plugins/libRegisterImagesModule.so: >> Wrote only 3584 of 10240 bytes >> tar: Skipping to next header >> tar: Archive contains obsolescent base-64 headers >> tar: Error exit delayed from previous errors >> >> I'm guessing this is the problem. Let me rebuild on Linux and see if that >> fixes it. >> >> Katie >> >> On Thu, 19 Mar 2009, Ron Kikinis wrote: >> >>> Clement, >>> Register images is included in the nightly build at least on the MAC. >>> Ron >>> >>> cvachet at email.unc.edu wrote: >>>> Dear all, >>>> >>>> When we compile Slicer3 from the source code, RegisterImages is >>>> compiled by default. However RegisterImages doesn't belong to Slicer3 >>>> nigthly builds. Is there a specific reason? In our regional cortical >>>> thickness analysis pipeline, ARCTIC (Slicer3 external module) runs >>>> several Slicer modules: RegisterImages, ResampleVolume2, ModelMaker. As >>>> we mostly use nightly builds versions of Slicer, we would be happy to >>>> have RegisterImages available. >>>> >>>> Thanks, >>>> Best regards, >>>> Clement Vachet. >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as >>>> the subject >>>> >>>> >>>> The information in this e-mail is intended only for the person to >>>> whom it is >>>> addressed. If you believe this e-mail was sent to you in error and >>>> the e-mail >>>> contains patient information, please contact the Partners >>>> Compliance HelpLine at >>>> http://www.partners.org/complianceline . If the e-mail was sent to >>>> you in error >>>> but does not contain patient information, please contact the >>>> sender and properly >>>> dispose of the e-mail. >>>> >>> >>> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as >> the subject >> > From daniel.blezek at gmail.com Thu Mar 19 21:03:27 2009 From: daniel.blezek at gmail.com (Daniel Blezek) Date: Thu, 19 Mar 2009 20:03:27 -0500 Subject: [slicer-devel] Commit problems In-Reply-To: <49C122DF.2080501@kitware.com> References: <1CC66820-5B12-4DD1-BC40-1E8DC3AC3379@gmail.com> <49C11E40.1010706@bwh.harvard.edu> <49C122DF.2080501@kitware.com> Message-ID: Strange, I was sure I deleted me entire Slicer-lib directory. I'll give it a go tonight. Thanks! -dan On Mar 18, 2009, at 11:35 AM, Zack Galbreath wrote: > Daniel Blezek wrote: >>> Is this a problem on my end? The code goes in: http://www.na-mic.org/svn/Slicer3-lib-mirrors/trunk/cmcurl >>> > > This is the old location of the SVN repository. It's still online > but nobody has write access to it anymore. The new URL is here: > > http://svn.slicer.org/Slicer3-lib-mirrors/trunk > > Here's an article on how to update your old checkout to point to the > new location of the repository: > > http://www.na-mic.org/Wiki/index.php/2008-Project-Week-SVN-Change > > -Zack -- Daniel Blezek daniel.blezek at gmail.com From nicole at bwh.harvard.edu Fri Mar 20 17:21:21 2009 From: nicole at bwh.harvard.edu (Nicole Aucoin) Date: Fri, 20 Mar 2009 17:21:21 -0400 (EDT) Subject: [slicer-devel] =?utf-8?q?Create_LabelMap_from_Model=3F?= In-Reply-To: <94f5b98593c12c2f85e0df4b3a5a907d@bwh.harvard.edu> References: <1237326316.5051.26.camel@thanks.bwh.harvard.edu> <94f5b98593c12c2f85e0df4b3a5a907d@bwh.harvard.edu> Message-ID: Hello, Does your cylinder model appear in the Data node's tree? The ModelIntoLabelVolume only works on model nodes that have been added to the scene. Have you tried using the Subdivide Model flag to generate more vertices? Are you seeing any error messages in the Log window? Also check in the standard output from the module execution. Can you send me a .vtk file of your cylinder model for me to try out? Nicole On Wed, 18 Mar 2009, Wolfgang Schramm wrote: > Hi Isaiah, > > thank you for your input. ModelIntoLabelVolume looks very promising, but it > doesn't work 100% for me. > It works fine with the sphere model, but it shows no result for my > cylinder. I assume my method of creating the cylinder is the cause for > this: > > 1) Create two points (I use fiducials for that purpose) > 2) Create a lineSource between these points > 3) Create the cylinder by feeding the linesource through a tube filter. > > Do you have any idea on how to circumvent this problem? > > Thank you > Wolfgang > > On Tue, 17 Mar 2009 17:45:16 -0400, Isaiah Norton > wrote: >> Hi Wolfgang, >> >> This module might help: Model Generation->Model_Into_Label_Volume. This >> colors voxels by polydata vertices so will give you the outline of the >> model (not the middle) >> >> The above was designed specifically to label tract-volume intersections >> so YMMV with other models (worked for a basic sphereish tumor model). It >> might be inefficient for a very dense model b/c it searches every >> vertex. >> >> I think the vtk function to do this is vtkImplicitFunction. >> >> -I >> >> >> On Tue, 2009-03-17 at 16:50 -0400, Wolfgang Schramm wrote: >>> Hello, >>> >>> I'm currently trying to create a labelmap from a 3D Model (simple sphere >>> and cylinder). Is there a way to do it in the Slicer UI, or can anybody >>> point me to the correct vtk/slicer classes to use? >>> >>> Thank you >>> Wolfgang >>> >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >>> >>> >>> The information in this e-mail is intended only for the person to whom >> it is >>> addressed. If you believe this e-mail was sent to you in error and the >> e-mail >>> contains patient information, please contact the Partners Compliance >> HelpLine at >>> http://www.partners.org/complianceline . If the e-mail was sent to you >> in error >>> but does not contain patient information, please contact the sender and >> properly >>> dispose of the e-mail. >>> >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > -- Nicole Aucoin nicole at bwh.harvard.edu Surgical Planning Laboratory, Brigham and Women's Hospital Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 http://www.slicer.org From wschramm at bwh.harvard.edu Sat Mar 21 15:04:58 2009 From: wschramm at bwh.harvard.edu (Wolfgang Schramm) Date: Sat, 21 Mar 2009 15:04:58 -0400 Subject: [slicer-devel] =?utf-8?q?Create_LabelMap_from_Model=3F?= In-Reply-To: References: Message-ID: <713b1fd0d404b37ece551f2fb525a64c@bwh.harvard.edu> Hi Nicole, Yes, the cylinder appears in the Data Tree (as a vtkMRMLModelNode). I tried with and without the subdivide flag, no difference in the result tough. Error Log is empty. Attached is the vtk file. thank you Wolfgang On Fri, 20 Mar 2009 17:21:21 -0400 (EDT), Nicole Aucoin wrote: > Hello, > > Does your cylinder model appear in the Data node's tree? The > ModelIntoLabelVolume only works on model nodes that have been added to the > scene. > Have you tried using the Subdivide Model flag to generate more vertices? > Are you seeing any error messages in the Log window? Also check in the > standard output from the module execution. > Can you send me a .vtk file of your cylinder model for me to try out? > > Nicole > > On Wed, 18 Mar 2009, Wolfgang Schramm wrote: > >> Hi Isaiah, >> >> thank you for your input. ModelIntoLabelVolume looks very promising, but > it >> doesn't work 100% for me. >> It works fine with the sphere model, but it shows no result for my >> cylinder. I assume my method of creating the cylinder is the cause for >> this: >> >> 1) Create two points (I use fiducials for that purpose) >> 2) Create a lineSource between these points >> 3) Create the cylinder by feeding the linesource through a tube filter. >> >> Do you have any idea on how to circumvent this problem? >> >> Thank you >> Wolfgang >> >> On Tue, 17 Mar 2009 17:45:16 -0400, Isaiah Norton >> wrote: >>> Hi Wolfgang, >>> >>> This module might help: Model Generation->Model_Into_Label_Volume. This >>> colors voxels by polydata vertices so will give you the outline of the >>> model (not the middle) >>> >>> The above was designed specifically to label tract-volume intersections >>> so YMMV with other models (worked for a basic sphereish tumor model). > It >>> might be inefficient for a very dense model b/c it searches every >>> vertex. >>> >>> I think the vtk function to do this is vtkImplicitFunction. >>> >>> -I >>> >>> >>> On Tue, 2009-03-17 at 16:50 -0400, Wolfgang Schramm wrote: >>>> Hello, >>>> >>>> I'm currently trying to create a labelmap from a 3D Model (simple > sphere >>>> and cylinder). Is there a way to do it in the Slicer UI, or can > anybody >>>> point me to the correct vtk/slicer classes to use? >>>> >>>> Thank you >>>> Wolfgang >>>> >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >>>> >>>> >>>> The information in this e-mail is intended only for the person to whom >>> it is >>>> addressed. If you believe this e-mail was sent to you in error and the >>> e-mail >>>> contains patient information, please contact the Partners Compliance >>> HelpLine at >>>> http://www.partners.org/complianceline . If the e-mail was sent to you >>> in error >>>> but does not contain patient information, please contact the sender > and >>> properly >>>> dispose of the e-mail. >>>> >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject >> > > -- > Nicole Aucoin nicole at bwh.harvard.edu > Surgical Planning Laboratory, Brigham and Women's Hospital > Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 > http://www.slicer.org -------------- next part -------------- A non-text attachment was scrubbed... Name: probe.vtk Type: application/octet-stream Size: 4946 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090321/cba691c7/attachment.obj From nicole at bwh.harvard.edu Mon Mar 23 12:22:43 2009 From: nicole at bwh.harvard.edu (Nicole Aucoin) Date: Mon, 23 Mar 2009 12:22:43 -0400 (EDT) Subject: [slicer-devel] =?utf-8?q?Create_LabelMap_from_Model=3F?= In-Reply-To: <713b1fd0d404b37ece551f2fb525a64c@bwh.harvard.edu> References: <713b1fd0d404b37ece551f2fb525a64c@bwh.harvard.edu> Message-ID: I think I see the problem: if you display the cylinder in the 3d view and hit 'w' to turn on wireframe rendering, it looks like the cylinder is defined by very few points, along the circumference of the end caps. I tried manually subdividing, and that filter actually fails on this data set (goes from 80 points to 0). Can you create the cylinder in a different way such that it has more points defined? There is a new module being worked upon that will do a flood fill and not rely so much on the vertices of models, it'll be available once the feature freeze is lifted from the Slicer svn repository. Nicole On Sat, 21 Mar 2009, Wolfgang Schramm wrote: > Hi Nicole, > > Yes, the cylinder appears in the Data Tree (as a vtkMRMLModelNode). I tried > with and without the subdivide flag, no difference in the result tough. > Error Log is empty. > > Attached is the vtk file. > > thank you > Wolfgang > > > On Fri, 20 Mar 2009 17:21:21 -0400 (EDT), Nicole Aucoin > wrote: >> Hello, >> >> Does your cylinder model appear in the Data node's tree? The >> ModelIntoLabelVolume only works on model nodes that have been added to > the >> scene. >> Have you tried using the Subdivide Model flag to generate more vertices? >> Are you seeing any error messages in the Log window? Also check in the >> standard output from the module execution. >> Can you send me a .vtk file of your cylinder model for me to try out? >> >> Nicole >> >> On Wed, 18 Mar 2009, Wolfgang Schramm wrote: >> >>> Hi Isaiah, >>> >>> thank you for your input. ModelIntoLabelVolume looks very promising, but >> it >>> doesn't work 100% for me. >>> It works fine with the sphere model, but it shows no result for my >>> cylinder. I assume my method of creating the cylinder is the cause for >>> this: >>> >>> 1) Create two points (I use fiducials for that purpose) >>> 2) Create a lineSource between these points >>> 3) Create the cylinder by feeding the linesource through a tube filter. >>> >>> Do you have any idea on how to circumvent this problem? >>> >>> Thank you >>> Wolfgang >>> >>> On Tue, 17 Mar 2009 17:45:16 -0400, Isaiah Norton >>> wrote: >>>> Hi Wolfgang, >>>> >>>> This module might help: Model Generation->Model_Into_Label_Volume. This >>>> colors voxels by polydata vertices so will give you the outline of the >>>> model (not the middle) >>>> >>>> The above was designed specifically to label tract-volume intersections >>>> so YMMV with other models (worked for a basic sphereish tumor model). >> It >>>> might be inefficient for a very dense model b/c it searches every >>>> vertex. >>>> >>>> I think the vtk function to do this is vtkImplicitFunction. >>>> >>>> -I >>>> >>>> >>>> On Tue, 2009-03-17 at 16:50 -0400, Wolfgang Schramm wrote: >>>>> Hello, >>>>> >>>>> I'm currently trying to create a labelmap from a 3D Model (simple >> sphere >>>>> and cylinder). Is there a way to do it in the Slicer UI, or can >> anybody >>>>> point me to the correct vtk/slicer classes to use? >>>>> >>>>> Thank you >>>>> Wolfgang >>>>> >>>>> >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>> subject >>>>> >>>>> >>>>> The information in this e-mail is intended only for the person to whom >>>> it is >>>>> addressed. If you believe this e-mail was sent to you in error and the >>>> e-mail >>>>> contains patient information, please contact the Partners Compliance >>>> HelpLine at >>>>> http://www.partners.org/complianceline . If the e-mail was sent to you >>>> in error >>>>> but does not contain patient information, please contact the sender >> and >>>> properly >>>>> dispose of the e-mail. >>>>> >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>> subject >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >>> >> >> -- >> Nicole Aucoin nicole at bwh.harvard.edu >> Surgical Planning Laboratory, Brigham and Women's Hospital >> Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 >> http://www.slicer.org -- Nicole Aucoin nicole at bwh.harvard.edu Surgical Planning Laboratory, Brigham and Women's Hospital Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 http://www.slicer.org From fedorov at bwh.harvard.edu Mon Mar 23 12:38:46 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 23 Mar 2009 12:38:46 -0400 Subject: [slicer-devel] Create LabelMap from Model? In-Reply-To: References: <713b1fd0d404b37ece551f2fb525a64c@bwh.harvard.edu> Message-ID: <82f536810903230938n69cd823bu45f15cade0865f97@mail.gmail.com> Wolfgang, I did not have a need to do something like this, but here's a seemingly very easy way to solve your problem with ITK tools: 1) create a cylinder spatial object with specified height and radius using ITK CylinderSpatialObject class http://www.itk.org/Doxygen/html/classitk_1_1CylinderSpatialObject.html 2) convert your spatial object to binary image with ITK SpatialObjectToImageFilter class http://www.itk.org/Doxygen/html/classitk_1_1SpatialObjectToImageFilter.html 3) save your binary image in a format of your choice using ITK ImageFileWriter http://www.itk.org/Doxygen/html/classitk_1_1SpatialObjectToImageFilter.html Unfortunately, you cannot use ITK images directly in Slicer. If you want to avoid communicating through files, see the discussion here http://wiki.na-mic.org/Wiki/index.php/2009_Winter_Project_Week_vtkITK_Pipeline for some pointers. You should be able to create spatial object and transform it into binary mask in ITK, and then use vtkITK magic to get a VTK image. Hope this helps Andriy Fedorov On Mon, Mar 23, 2009 at 12:22 PM, Nicole Aucoin wrote: > I think I see the problem: if you display the cylinder in the 3d view and > hit 'w' to turn on wireframe rendering, it looks like the cylinder is > defined by very few points, along the circumference of the end caps. I > tried manually subdividing, and that filter actually fails on this data > set (goes from 80 points to 0). > > Can you create the cylinder in a different way such that it has more > points defined? > > There is a new module being worked upon that will do a flood fill and not > rely so much on the vertices of models, it'll be available once the > feature freeze is lifted from the Slicer svn repository. > > Nicole > > On Sat, 21 Mar 2009, Wolfgang Schramm wrote: > >> Hi Nicole, >> >> Yes, the cylinder appears in the Data Tree (as a vtkMRMLModelNode). I tried >> with and without the subdivide flag, no difference in the result tough. >> Error Log is empty. >> >> Attached is the vtk file. >> >> thank you >> Wolfgang >> >> >> On Fri, 20 Mar 2009 17:21:21 -0400 (EDT), Nicole Aucoin >> wrote: >>> Hello, >>> >>> Does your cylinder model appear in the Data node's tree? The >>> ModelIntoLabelVolume only works on model nodes that have been added to >> the >>> scene. >>> Have you tried using the Subdivide Model flag to generate more vertices? >>> Are you seeing any error messages in the Log window? Also check in the >>> standard output from the module execution. >>> Can you send me a .vtk file of your cylinder model for me to try out? >>> >>> Nicole >>> >>> On Wed, 18 Mar 2009, Wolfgang Schramm wrote: >>> >>>> Hi Isaiah, >>>> >>>> thank you for your input. ModelIntoLabelVolume looks very promising, but >>> it >>>> doesn't work 100% for me. >>>> It works fine with the sphere model, but it shows no result for my >>>> cylinder. I assume my method of creating the cylinder is the cause for >>>> this: >>>> >>>> 1) Create two points (I use fiducials for that purpose) >>>> 2) Create a lineSource between these points >>>> 3) Create the cylinder by feeding the linesource through a tube filter. >>>> >>>> Do you have any idea on how to circumvent this problem? >>>> >>>> Thank you >>>> Wolfgang >>>> >>>> On Tue, 17 Mar 2009 17:45:16 -0400, Isaiah Norton >>>> wrote: >>>>> Hi Wolfgang, >>>>> >>>>> This module might help: Model Generation->Model_Into_Label_Volume. This >>>>> colors voxels by polydata vertices so will give you the outline of the >>>>> model (not the middle) >>>>> >>>>> The above was designed specifically to label tract-volume intersections >>>>> so YMMV with other models (worked for a basic sphereish tumor model). >>> It >>>>> might be inefficient for a very dense model b/c it searches every >>>>> vertex. >>>>> >>>>> I think the vtk function to do this is vtkImplicitFunction. >>>>> >>>>> -I >>>>> >>>>> >>>>> On Tue, 2009-03-17 at 16:50 -0400, Wolfgang Schramm wrote: >>>>>> Hello, >>>>>> >>>>>> I'm currently trying to create a labelmap from a 3D Model (simple >>> sphere >>>>>> and cylinder). Is there a way to do it in the Slicer UI, or can >>> anybody >>>>>> point me to the correct vtk/slicer classes to use? >>>>>> >>>>>> Thank you >>>>>> Wolfgang >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> slicer-devel mailing list >>>>>> slicer-devel at bwh.harvard.edu >>>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>>> subject >>>>>> >>>>>> >>>>>> The information in this e-mail is intended only for the person to whom >>>>> it is >>>>>> addressed. If you believe this e-mail was sent to you in error and the >>>>> e-mail >>>>>> contains patient information, please contact the Partners Compliance >>>>> HelpLine at >>>>>> http://www.partners.org/complianceline . If the e-mail was sent to you >>>>> in error >>>>>> but does not contain patient information, please contact the sender >>> and >>>>> properly >>>>>> dispose of the e-mail. >>>>>> >>>>> >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>>> subject >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >>>> >>> >>> -- >>> Nicole Aucoin ? ? ? ? ? ? ? ? ? ? ? nicole at bwh.harvard.edu >>> Surgical Planning Laboratory, Brigham and Women's Hospital >>> Rm 206, 1249 Boylston St., Boston, MA 02215 ? 617-525-6216 >>> ? ? ? ? ? ? ? http://www.slicer.org > > -- > Nicole Aucoin ? ? ? ? ? ? ? ? ? ? ? nicole at bwh.harvard.edu > Surgical Planning Laboratory, Brigham and Women's Hospital > Rm 206, 1249 Boylston St., Boston, MA 02215 ? 617-525-6216 > ? ? ? ? ? ? ?http://www.slicer.org > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From vrnova at gmail.com Mon Mar 23 12:49:26 2009 From: vrnova at gmail.com (Yanling Liu) Date: Mon, 23 Mar 2009 12:49:26 -0400 Subject: [slicer-devel] KWWidgets New and Delete Message-ID: <24ceacd90903230949j2bc9920aqa12090b9dc9a18d6@mail.gmail.com> Hello, I need a little hint about KWWidgets::New and Delete functions. Whenever I created a new KWWigets widget, do I need to manually call the Delete() function for the new widget? Is KWWidgets smart enough to automatically delete instances? Thanks! Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090323/c6c216a4/attachment.html From fedorov at bwh.harvard.edu Mon Mar 23 13:03:57 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 23 Mar 2009 13:03:57 -0400 Subject: [slicer-devel] KWWidgets New and Delete In-Reply-To: <24ceacd90903230949j2bc9920aqa12090b9dc9a18d6@mail.gmail.com> References: <24ceacd90903230949j2bc9920aqa12090b9dc9a18d6@mail.gmail.com> Message-ID: <82f536810903231003i4419fc33we91a13a2c0a81134@mail.gmail.com> Yanling, Yes, you need to explicitly deallocate KWWidget objects allocated with New(), to the best of my knowledge, only ITK objects are using smart pointers and need not be explicitly deallocated. An alternative is to use SmartPointers, as discussed here http://slicer.spl.harvard.edu/slicerWiki/index.php/Slicer3:Memory_Management, but if you look at Slicer code, they are used very infrequently compared to New/Delete's. Maybe someone else can comment why this is the case (possibly, because it is a relatively new addition to Slicer). Fedorov On Mon, Mar 23, 2009 at 12:49 PM, Yanling Liu wrote: > Hello, I need a little hint about KWWidgets::New and Delete functions. > Whenever I created a new KWWigets widget, do I need to manually call the > Delete() function for the new widget? Is KWWidgets smart enough to > automatically delete instances? > > Thanks! > Yanling > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > > > The information in this e-mail is intended only for the person to whom it is > addressed. If you believe this e-mail was sent to you in error and the > e-mail > contains patient information, please contact the Partners Compliance > HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent to you in > error > but does not contain patient information, please contact the sender and > properly > dispose of the e-mail. > > From norman-k-williams at uiowa.edu Mon Mar 23 13:26:23 2009 From: norman-k-williams at uiowa.edu (kent williams) Date: Mon, 23 Mar 2009 12:26:23 -0500 Subject: [slicer-devel] KWWidgets New and Delete In-Reply-To: <24ceacd90903230949j2bc9920aqa12090b9dc9a18d6@mail.gmail.com> Message-ID: That?s a good question perhaps better suited to the KWWidgets user list. But here is my understanding of the issue: All KWWidgets classes are derived ultimately from vtkObjectBase, which supports reference counting. When you call New() for a KWWidgets class, it has a reference count of one. That means it needs balancing with a Delete() -- usually. There?s a page on the Slicer WIKI that explains this pretty well: http://www.slicer.org/slicerWiki/index.php/Slicer3:Memory_Management VTK Memory management is more complicated than is described in that WIKI page*, but not so much in ways that would affect using KWWidgets. I?d say this is the ?best case? discipline to follow: 1. Instead of using raw pointers to KWWIdgets as member variables in your KWWidgets classes, use vtkSmartPointers. 2. Assign results of New() to the smart pointers. This will mean that when your class instance is destroyed, the smart pointer members are also destroyed, and they do the right thing with respect to freeing memory and resources. 3. Where possible use smart pointers in preference to raw pointers for local variables in a function. This means that when the smart pointer goes out of scope, it?s automatically destroyed, and the right thing happens to the object pointed to. *the complication arises when you create vtk Data obects, and then pass them into other vtk methods. I don?t know what the rule is, or if VTK is 100% consistent about this, but some times, passing a VTK object to a VTK method transfers ownership of the pointer, so calling Delete() after calling some methods is in fact an error. On 3/23/09 11:49 AM, "Yanling Liu" wrote: > Hello, I need a little hint about KWWidgets::New and Delete functions. > Whenever I created a new KWWigets widget, do I need to manually call the > Delete() function for the new widget? Is KWWidgets smart enough to > automatically delete instances? > > Thanks! > Yanling > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged. If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited. Please reply to the sender that you have received the message in error, then delete it. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090323/ddf2f4de/attachment.html From simon at math.uh.edu Thu Mar 26 12:09:52 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Thu, 26 Mar 2009 11:09:52 -0500 Subject: [slicer-devel] strange interaction, slicer & KDE Message-ID: <200903261109.52457.simon@math.uh.edu> I've found a very strange bug, but I'm exactly sure where the fault lies. I'd appreciate any insight anyone has I've been using Slicer3 from trunk builds on both OS X and a KDE based linux system. Recently the IT folks here upgraded the KDE system (to 3.5). Around the same time I noticed something very strange with my Slicer3 builds (I can't be certain that it was precisely when the change happened, unfortunately). Here's the bug: Whenever Slicer3 was started, during the initialization the mouse pointer would change to a crosshair and the system bell would bleep a few times. Initialization was slow. I've managed to narrow down what was going on. During initialization, some snapshot routine is being triggered and postscript screenshots are being taken of the active window at the time. After finding these files, I realized they were always named after python modules (is that the right name? I'm brand-new to python) like "os", "time", "numpy" (that latter one gave it away). Digging down a little more, I've discovered that the weird interaction occurs when Slicer loads a python plugin that has a top level import statement. To be explicit: On my system, if you make a file like this this: > XML = """ > > Filtering > trigger snapshot bug > python module > 0.1 > > simon > > """ > > import os > > def Execute (): > > pass And you put this file in Slicer3-build/lib/Slicer3/Plugins, then the system will take a snapshot (when you press a mouse buttton) of the selected window and save it as postscript in a file called "os". Get rid of the import statement and the problem goes away. Weird, isn't it? This doesn't happen with scripts in child directories of Plugins. It doesn't happen with Modules, either. My first hypothesis is that the Slicer/python process is somehow sending control characters to the window manager that have been bound to a "snapshot". So far (I'm not very familiar with KDE linux) I haven't found such a process or binding, but I'm not sure where to look. The problem doesn't occur on my OS X machines. Has anyone else seen similar behavior, or have some insight? It's mixing a lot of things I hardly know at all (python, KDE, Slicer3s load process) so I may be missing an obvious clue. cheers, simon ps: I can work around it by removing the half dozen or so scripts that have top level import statements, but that's hardly a long term solution. From simon at math.uh.edu Thu Mar 26 12:14:47 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Thu, 26 Mar 2009 11:14:47 -0500 Subject: [slicer-devel] addendum to previous message about triggering screenshot bug Message-ID: <200903261114.47645.simon@math.uh.edu> I should have added that, whatever is going on it is *not* the typical screen capture method in the system, that one (triggered by a print screen or directly) will open a dialog to capture as .png. I also forgot to ask if there is any such postscript screenshot capability in Slicer somewhere... cheers, simon. From simon at math.uh.edu Thu Mar 26 12:18:58 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Thu, 26 Mar 2009 11:18:58 -0500 Subject: [slicer-devel] sorry to triple post (again on the screenshots) Message-ID: <200903261118.58758.simon@math.uh.edu> I had originally made a mistake looking at the postscript. I thought it wasn't tagged, but it turns out the header does tell me that ImageMagick made the postscript files. Don't know if that helps or not, as it is widely used as a library. simon From demian.wassermann at sophia.inria.fr Thu Mar 26 13:59:47 2009 From: demian.wassermann at sophia.inria.fr (Demian Wassermann) Date: Thu, 26 Mar 2009 18:59:47 +0100 Subject: [slicer-devel] strange interaction, slicer & KDE In-Reply-To: <200903261109.52457.simon@math.uh.edu> References: <200903261109.52457.simon@math.uh.edu> Message-ID: <71226E42-3878-4B45-AEBB-5F4D83F15714@sophia.inria.fr> Hello Simon, Are you using your system python or the one built by slicer? Do you have any python environment variables on the system? could you run your python and enter the instructions import os exit() and check if there's a snapshot? Demian -- Demian Wassermann demian.wassermann at sophia.inria.fr PhD Student Odyssee Research Project INRIA Sophia-Antipolis 2004 route des lucioles - FR-06902 On Mar 26, 2009, at 5:09 PM, Simon K Alexander wrote: > I've found a very strange bug, but I'm exactly sure where the fault > lies. I'd appreciate any insight anyone has > > I've been using Slicer3 from trunk builds on both OS X and a KDE > based linux system. > > Recently the IT folks here upgraded the KDE system (to 3.5). > Around the same time I noticed something very strange with > my Slicer3 builds (I can't be certain that it was precisely when > the change happened, unfortunately). > > Here's the bug: Whenever Slicer3 was started, during the > initialization the mouse pointer would change to a crosshair > and the system bell would bleep a few times. Initialization was > slow. > > I've managed to narrow down what was going on. During > initialization, some snapshot routine is being triggered and > postscript screenshots are being taken of the active window at the > time. After finding these files, I realized they > were always named after python modules (is that the right name? > I'm brand-new to python) like "os", "time", "numpy" > (that latter one gave it away). > > Digging down a little more, I've discovered that the weird > interaction occurs when Slicer loads a python plugin that has > a top level import statement. > > To be explicit: On my system, if you make a file like this this: > >> XML = """ >> >> Filtering >> trigger snapshot bug >> python module >> 0.1 >> >> simon >> >> """ >> >> import os >> >> def Execute (): >> >> pass > > > And you put this file in Slicer3-build/lib/Slicer3/Plugins, then > the system will take a snapshot (when you press a > mouse buttton) of the selected window and save it as postscript in > a file called "os". Get rid of the import statement > and the problem goes away. > > Weird, isn't it? > > > This doesn't happen with scripts in child directories of Plugins. > It doesn't happen with Modules, either. > > My first hypothesis is that the Slicer/python process is somehow > sending control characters to the window manager that > have been bound to a "snapshot". So far (I'm not very familiar > with KDE linux) I haven't found such a process or > binding, but I'm not sure where to look. The problem doesn't occur > on my OS X machines. > > > Has anyone else seen similar behavior, or have some insight? It's > mixing a lot of things I hardly know at all (python, > KDE, Slicer3s load process) so I may be missing an obvious clue. > > cheers, > simon > > ps: I can work around it by removing the half dozen or so scripts > that have top level import statements, but that's > hardly a long term solution. > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel- > request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Thu Mar 26 14:01:17 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 26 Mar 2009 14:01:17 -0400 Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <200903261118.58758.simon@math.uh.edu> References: <200903261118.58758.simon@math.uh.edu> Message-ID: <49CBC2ED.4060707@bwh.harvard.edu> Hi Simon - Very interesting - thanks for the detective work! I don't know the fix yet but I have an idea what's going on. I started seeing the cursor change behavior also on mac a couple days ago but didn't understand it until now. I think the python code is being run as if it were a shell script so when the "import numpy" line shows up in a file, it resolves on my machine to run the /opt/local/bin/import with numpy as an argument. I'll take a closer look... -Steve Simon K Alexander wrote: > I had originally made a mistake looking at the postscript. I thought it wasn't tagged, but it turns out the header does > tell me that ImageMagick made the postscript files. > > Don't know if that helps or not, as it is widely used as a library. > > simon > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From curtislisle at knowledgevis.com Thu Mar 26 14:44:39 2009 From: curtislisle at knowledgevis.com (curtislisle at knowledgevis.com) Date: Thu, 26 Mar 2009 14:44:39 -0400 (EDT) Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <49CBC2ED.4060707@bwh.harvard.edu> References: <200903261118.58758.simon@math.uh.edu> <49CBC2ED.4060707@bwh.harvard.edu> Message-ID: <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> I am experiencing similar Slicer3 startup issues (cursor as a cross, repeated beeps during startup) on a CentOS 5 system running Gnome. Slicer doesn't complain on the console or error log during startup and it functions correctly. Slicer revision: 8969 (downloaded yesterday) [root at linux-vrc bin]# uname -a Linux linux-vrc 2.6.18-92.1.22.el5 #1 SMP Tue Dec 16 11:57:43 EST 2008 x86_64 x86_64 x86_64 GNU/Linux Curt On Thu, March 26, 2009 2:01 pm, Steve Pieper wrote: > Hi Simon - > > > Very interesting - thanks for the detective work! I don't know the fix > yet but I have an idea what's going on. I started seeing the cursor change > behavior also on mac a couple days ago but didn't understand it until now. > > > I think the python code is being run as if it were a shell script so > when the "import numpy" line shows up in a file, it resolves on my machine > to run the /opt/local/bin/import with numpy as an argument. > > I'll take a closer look... > > > -Steve > > > Simon K Alexander wrote: > >> I had originally made a mistake looking at the postscript. I thought >> it wasn't tagged, but it turns out the header does tell me that >> ImageMagick made the postscript files. >> >> >> Don't know if that helps or not, as it is widely used as a library. >> >> >> simon _______________________________________________ >> slicer-devel mailing list slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > _______________________________________________ > slicer-devel mailing list slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > From luca.antiga at gmail.com Fri Mar 27 12:05:45 2009 From: luca.antiga at gmail.com (Luca Antiga) Date: Fri, 27 Mar 2009 17:05:45 +0100 Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> References: <200903261118.58758.simon@math.uh.edu> <49CBC2ED.4060707@bwh.harvard.edu> <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> Message-ID: <2d509e140903270905p7912357fr4288824ae83ec23e@mail.gmail.com> Hi guys, Steve is right, this happens when the Python files have executable permissions and there's no #!/usr/bin/env python at the beginning of the file, so they are executed using the bash interpreter, which in turn calls import which is the screenshot acquisition executable from Image Magick (I always wondered who had the idea of calling Image Magick executables 'convert', 'import', 'identify', 'display'... they should be given a fine for this!) I haven't tried myself (I'm currently at a meeting) but the problem should be fixed by 1. making sure the python file starts with #!/usr/bin/env python 2. removing executable permissions 3. eventually avoiding to try to execute .py files during module discovery, although this is a double-edged knife, since it would prevent real executables written in python and having a py extension to be discovered. Cheers! Luca On Thu, Mar 26, 2009 at 7:44 PM, wrote: > I am experiencing similar Slicer3 startup issues (cursor as a cross, > repeated beeps during startup) on a CentOS 5 system running Gnome. Slicer > doesn't complain on the console or error log during startup and it > functions correctly. > > Slicer revision: 8969 (downloaded yesterday) > > [root at linux-vrc bin]# uname -a > Linux linux-vrc 2.6.18-92.1.22.el5 #1 SMP Tue Dec 16 11:57:43 EST 2008 > x86_64 x86_64 x86_64 GNU/Linux > > Curt > > > On Thu, March 26, 2009 2:01 pm, Steve Pieper wrote: > > Hi Simon - > > > > > > Very interesting - thanks for the detective work! I don't know the fix > > yet but I have an idea what's going on. I started seeing the cursor > change > > behavior also on mac a couple days ago but didn't understand it until > now. > > > > > > I think the python code is being run as if it were a shell script so > > when the "import numpy" line shows up in a file, it resolves on my > machine > > to run the /opt/local/bin/import with numpy as an argument. > > > > I'll take a closer look... > > > > > > -Steve > > > > > > Simon K Alexander wrote: > > > >> I had originally made a mistake looking at the postscript. I thought > >> it wasn't tagged, but it turns out the header does tell me that > >> ImageMagick made the postscript files. > >> > >> > >> Don't know if that helps or not, as it is widely used as a library. > >> > >> > >> simon _______________________________________________ > >> slicer-devel mailing list slicer-devel at bwh.harvard.edu > >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > >> To unsubscribe: send email to > >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > >> subject > > _______________________________________________ > > slicer-devel mailing list slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > > subject > > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > -- Luca Antiga, PhD Biomedical Technologies Laboratory Biomedical Engineering Department, Mario Negri Institute mail: Villa Camozzi, 24020, Ranica (BG), Italy phone: +39 035 4535-381 email: antiga at marionegri.it web: http://villacamozzi.marionegri.it/~luca -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090327/6c8030c0/attachment.html From simon at math.uh.edu Fri Mar 27 14:21:36 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Fri, 27 Mar 2009 13:21:36 -0500 Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <2d509e140903270905p7912357fr4288824ae83ec23e@mail.gmail.com> References: <200903261118.58758.simon@math.uh.edu> <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> <2d509e140903270905p7912357fr4288824ae83ec23e@mail.gmail.com> Message-ID: <200903271321.37239.simon@math.uh.edu> Hi Luca, Changing the executable status of these files in the build directory seems to clear everything up. However, the originals are not exectable, so the build process is setting this somehow. As you note, an executable might be actually wanted though, so something will have to be changed in the build to do the right thing here. About your 1st suggestion though: > 1. making sure the python file starts with #!/usr/bin/env python All of these files start with the XML= " ..." string. Does anything potentially rely on that? cheers, Simon On Friday 27 March 2009 11:05:45 am Luca Antiga wrote: > Hi guys, > Steve is right, this happens when the Python files have executable > permissions and there's no > #!/usr/bin/env python > at the beginning of the file, so they are executed using the bash > interpreter, which in turn calls > import > which is the screenshot acquisition executable from Image Magick (I always > wondered who had > the idea of calling Image Magick executables 'convert', 'import', > 'identify', 'display'... they should > be given a fine for this!) > I haven't tried myself (I'm currently at a meeting) but the problem should > be fixed by > 1. making sure the python file starts with #!/usr/bin/env python > 2. removing executable permissions > 3. eventually avoiding to try to execute .py files during module discovery, > although this is a double-edged > knife, since it would prevent real executables written in python and having > a py extension to be discovered. > > Cheers! > > Luca > > From simon at math.uh.edu Fri Mar 27 14:36:08 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Fri, 27 Mar 2009 13:36:08 -0500 Subject: [slicer-devel] ExecutionModelTourTest Message-ID: <200903271336.09109.simon@math.uh.edu> I've noticed recently that when doing a trunk build and test via getbuildtest.tcl, the file: Testing/Data/Input/ExecutionModelTourTest.mrml is overwritten in the source directory by a version with slightly different white space. Seems a bit strange to me. Any particular reason for this behavior? From pieper at bwh.harvard.edu Fri Mar 27 15:18:08 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 27 Mar 2009 15:18:08 -0400 Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <200903271321.37239.simon@math.uh.edu> References: <200903261118.58758.simon@math.uh.edu> <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> <2d509e140903270905p7912357fr4288824ae83ec23e@mail.gmail.com> <200903271321.37239.simon@math.uh.edu> Message-ID: <49CD2670.6030904@bwh.harvard.edu> Hi - We found which files were involved with this 'import' issue and they should be fixed pretty soon. In general we do need to be very careful about what ends up in the Plugins directory, since slicer will try to execute them with a '--xml' argument as part of the discovery process - so any unintended side effects could be very bad indeed... That's why they are in the special plugins directory rather than, for example, just using the PATH variable. -Steve p.s. (As Luca points out, the ImageMagick utilities are very poorly named - on windows there's another program named 'convert' that switches your disk from fat32 to ntfs! Kind of an unfriendly accident when you intended to convert an image to jpeg...). Simon K Alexander wrote: > Hi Luca, > > Changing the executable status of these files in the build directory seems to clear everything up. However, the > originals are not exectable, so the build process is setting this somehow. As you note, an executable might be > actually wanted though, so something will have to be changed in the build to do the right thing here. > > About your 1st suggestion though: >> 1. making sure the python file starts with #!/usr/bin/env python > > > All of these files start with the XML= " ..." string. Does anything potentially rely on that? > > cheers, > Simon > > On Friday 27 March 2009 11:05:45 am Luca Antiga wrote: >> Hi guys, >> Steve is right, this happens when the Python files have executable >> permissions and there's no >> #!/usr/bin/env python >> at the beginning of the file, so they are executed using the bash >> interpreter, which in turn calls >> import >> which is the screenshot acquisition executable from Image Magick (I always >> wondered who had >> the idea of calling Image Magick executables 'convert', 'import', >> 'identify', 'display'... they should >> be given a fine for this!) >> I haven't tried myself (I'm currently at a meeting) but the problem should >> be fixed by >> 1. making sure the python file starts with #!/usr/bin/env python >> 2. removing executable permissions >> 3. eventually avoiding to try to execute .py files during module discovery, >> although this is a double-edged >> knife, since it would prevent real executables written in python and having >> a py extension to be discovered. >> >> Cheers! >> >> Luca >> >> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From luca.antiga at gmail.com Sat Mar 28 12:46:58 2009 From: luca.antiga at gmail.com (Luca Antiga) Date: Sat, 28 Mar 2009 17:46:58 +0100 Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <49CD2670.6030904@bwh.harvard.edu> References: <200903261118.58758.simon@math.uh.edu> <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> <2d509e140903270905p7912357fr4288824ae83ec23e@mail.gmail.com> <200903271321.37239.simon@math.uh.edu> <49CD2670.6030904@bwh.harvard.edu> Message-ID: <2d509e140903280946g221ab0f5k4bdc8da744f22fb1@mail.gmail.com> Hey, did not know about the convert on Windows, pretty funny. Since this happened, do you think it would be worth to change the location of Python XML modules, e.g. to Plugins/Python or even Modules/Python to make the system more robust? I see this could happen again in the future, especially with local user's code. Luca 2009/3/27, Steve Pieper : > Hi - > > We found which files were involved with this 'import' issue and they > should be fixed pretty soon. > > In general we do need to be very careful about what ends up in the > Plugins directory, since slicer will try to execute them with a '--xml' > argument as part of the discovery process - so any unintended side > effects could be very bad indeed... That's why they are in the special > plugins directory rather than, for example, just using the PATH variable. > > -Steve > > p.s. (As Luca points out, the ImageMagick utilities are very poorly > named - on windows there's another program named 'convert' that switches > your disk from fat32 to ntfs! Kind of an unfriendly accident when you > intended to convert an image to jpeg...). > > > > Simon K Alexander wrote: >> Hi Luca, >> >> Changing the executable status of these files in the build directory seems >> to clear everything up. However, the >> originals are not exectable, so the build process is setting this somehow. >> As you note, an executable might be >> actually wanted though, so something will have to be changed in the build >> to do the right thing here. >> >> About your 1st suggestion though: >>> 1. making sure the python file starts with #!/usr/bin/env python >> >> >> All of these files start with the XML= " ..." string. Does anything >> potentially rely on that? >> >> cheers, >> Simon >> >> On Friday 27 March 2009 11:05:45 am Luca Antiga wrote: >>> Hi guys, >>> Steve is right, this happens when the Python files have executable >>> permissions and there's no >>> #!/usr/bin/env python >>> at the beginning of the file, so they are executed using the bash >>> interpreter, which in turn calls >>> import >>> which is the screenshot acquisition executable from Image Magick (I >>> always >>> wondered who had >>> the idea of calling Image Magick executables 'convert', 'import', >>> 'identify', 'display'... they should >>> be given a fine for this!) >>> I haven't tried myself (I'm currently at a meeting) but the problem >>> should >>> be fixed by >>> 1. making sure the python file starts with #!/usr/bin/env python >>> 2. removing executable permissions >>> 3. eventually avoiding to try to execute .py files during module >>> discovery, >>> although this is a double-edged >>> knife, since it would prevent real executables written in python and >>> having >>> a py extension to be discovered. >>> >>> Cheers! >>> >>> Luca >>> >>> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > -- Inviato dal mio dispositivo mobile Luca Antiga, PhD Biomedical Technologies Laboratory Biomedical Engineering Department, Mario Negri Institute mail: Villa Camozzi, 24020, Ranica (BG), Italy phone: +39 035 4535-381 email: antiga at marionegri.it web: http://villacamozzi.marionegri.it/~luca From fedorov at bwh.harvard.edu Mon Mar 30 19:01:29 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 30 Mar 2009 19:01:29 -0400 Subject: [slicer-devel] Slice translation in slice viewer Message-ID: <82f536810903301601v4551ba74p8c35e6a1cf26a758@mail.gmail.com> Hi, I was told that when the middle mouse button is pressed, the slice in slice viewer should translate following the mouse pointer. This functionality doesn't work on my platform (reported in http://na-mic.org/Mantis/view.php?id=460). Is there any workaround to re-center/translate slice views in Slicer? I know I can create a transform or change image origin, but I would rather not do that, because I need this only for visualization. Any hint is appreciated. Andriy Fedorov From pieper at bwh.harvard.edu Tue Mar 31 08:49:54 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 31 Mar 2009 08:49:54 -0400 Subject: [slicer-devel] Slice translation in slice viewer In-Reply-To: <82f536810903301601v4551ba74p8c35e6a1cf26a758@mail.gmail.com> References: <82f536810903301601v4551ba74p8c35e6a1cf26a758@mail.gmail.com> Message-ID: <49D21172.9090508@bwh.harvard.edu> Hi Andriy - We may need to add a workaround for that. Does the middle mouse work correctly elsewhere? I.e. can you translate the 3D viewer in slicer? Can you paste in an xterm? -Steve Andriy Fedorov wrote: > Hi, > > I was told that when the middle mouse button is pressed, the slice in > slice viewer should translate following the mouse pointer. This > functionality doesn't work on my platform (reported in > http://na-mic.org/Mantis/view.php?id=460). > > Is there any workaround to re-center/translate slice views in Slicer? > I know I can create a transform or change image origin, but I would > rather not do that, because I need this only for visualization. > > Any hint is appreciated. > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From fedorov at bwh.harvard.edu Tue Mar 31 09:13:06 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 31 Mar 2009 09:13:06 -0400 Subject: [slicer-devel] Slice translation in slice viewer In-Reply-To: <49D21172.9090508@bwh.harvard.edu> References: <82f536810903301601v4551ba74p8c35e6a1cf26a758@mail.gmail.com> <49D21172.9090508@bwh.harvard.edu> Message-ID: <82f536810903310613q62165ac7u35fd0818fa12ba86@mail.gmail.com> Steve, Yes, the middle button is functional otherwise (i.e., I can paste in an xterm). But not in Slicer -- in 3D viewer middle button does zoom, not translation. It is interesting that I have the same behavior with both the the external mouse: right button zoom, middle button scroll through slices. Maybe it is just something specific to my system setup, but I didn't do anything special. It would be nice if there was more than one way to do slice translation, in any case. Andriy Fedorov On Tue, Mar 31, 2009 at 8:49 AM, Steve Pieper wrote: > Hi Andriy - > > We may need to add a workaround for that. ?Does the middle mouse work > correctly elsewhere? ?I.e. can you translate the 3D viewer in slicer? Can > you paste in an xterm? > > -Steve > > Andriy Fedorov wrote: >> >> Hi, >> >> I was told that when the middle mouse button is pressed, the slice in >> slice viewer should translate following the mouse pointer. This >> functionality doesn't work on my platform (reported in >> http://na-mic.org/Mantis/view.php?id=460). >> >> Is there any workaround to re-center/translate slice views in Slicer? >> I know I can create a transform or change image origin, but I would >> rather not do that, because I need this only for visualization. >> >> Any hint is appreciated. >> >> Andriy Fedorov >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject > From wschramm at bwh.harvard.edu Wed Apr 1 10:17:28 2009 From: wschramm at bwh.harvard.edu (Wolfgang Schramm) Date: Wed, 1 Apr 2009 10:17:28 -0400 Subject: [slicer-devel] =?utf-8?q?Create_LabelMap_from_Model=3F?= In-Reply-To: References: Message-ID: <98fea53814c8549907c518b07403458c@bwh.harvard.edu> Thank you for your analysis - it's about what I've figured out after using the model with paraview. After some more internet research I stumbled upon vtkPolyDataToImageStencil - it looks promising, but since since I'm fairly new to vtk I'm not exactly sure on how to use it with a DICOM Volume. Has anybody already used it and can confirm the suitability for this purpose? Ty Wolfgang On Mon, 23 Mar 2009 12:22:43 -0400 (EDT), Nicole Aucoin wrote: > I think I see the problem: if you display the cylinder in the 3d view and > hit 'w' to turn on wireframe rendering, it looks like the cylinder is > defined by very few points, along the circumference of the end caps. I > tried manually subdividing, and that filter actually fails on this data > set (goes from 80 points to 0). > > Can you create the cylinder in a different way such that it has more > points defined? > > There is a new module being worked upon that will do a flood fill and not > rely so much on the vertices of models, it'll be available once the > feature freeze is lifted from the Slicer svn repository. > > Nicole > > On Sat, 21 Mar 2009, Wolfgang Schramm wrote: > >> Hi Nicole, >> >> Yes, the cylinder appears in the Data Tree (as a vtkMRMLModelNode). I > tried >> with and without the subdivide flag, no difference in the result tough. >> Error Log is empty. >> >> Attached is the vtk file. >> >> thank you >> Wolfgang >> >> >> On Fri, 20 Mar 2009 17:21:21 -0400 (EDT), Nicole Aucoin >> wrote: >>> Hello, >>> >>> Does your cylinder model appear in the Data node's tree? The >>> ModelIntoLabelVolume only works on model nodes that have been added to >> the >>> scene. >>> Have you tried using the Subdivide Model flag to generate more > vertices? >>> Are you seeing any error messages in the Log window? Also check in the >>> standard output from the module execution. >>> Can you send me a .vtk file of your cylinder model for me to try out? >>> >>> Nicole >>> >>> On Wed, 18 Mar 2009, Wolfgang Schramm wrote: >>> >>>> Hi Isaiah, >>>> >>>> thank you for your input. ModelIntoLabelVolume looks very promising, > but >>> it >>>> doesn't work 100% for me. >>>> It works fine with the sphere model, but it shows no result for my >>>> cylinder. I assume my method of creating the cylinder is the cause for >>>> this: >>>> >>>> 1) Create two points (I use fiducials for that purpose) >>>> 2) Create a lineSource between these points >>>> 3) Create the cylinder by feeding the linesource through a tube > filter. >>>> >>>> Do you have any idea on how to circumvent this problem? >>>> >>>> Thank you >>>> Wolfgang >>>> >>>> On Tue, 17 Mar 2009 17:45:16 -0400, Isaiah Norton > >>>> wrote: >>>>> Hi Wolfgang, >>>>> >>>>> This module might help: Model Generation->Model_Into_Label_Volume. > This >>>>> colors voxels by polydata vertices so will give you the outline of > the >>>>> model (not the middle) >>>>> >>>>> The above was designed specifically to label tract-volume > intersections >>>>> so YMMV with other models (worked for a basic sphereish tumor model). >>> It >>>>> might be inefficient for a very dense model b/c it searches every >>>>> vertex. >>>>> >>>>> I think the vtk function to do this is vtkImplicitFunction. >>>>> >>>>> -I >>>>> >>>>> >>>>> On Tue, 2009-03-17 at 16:50 -0400, Wolfgang Schramm wrote: >>>>>> Hello, >>>>>> >>>>>> I'm currently trying to create a labelmap from a 3D Model (simple >>> sphere >>>>>> and cylinder). Is there a way to do it in the Slicer UI, or can >>> anybody >>>>>> point me to the correct vtk/slicer classes to use? >>>>>> >>>>>> Thank you >>>>>> Wolfgang >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> slicer-devel mailing list >>>>>> slicer-devel at bwh.harvard.edu >>>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>>> subject >>>>>> >>>>>> >>>>>> The information in this e-mail is intended only for the person to > whom >>>>> it is >>>>>> addressed. If you believe this e-mail was sent to you in error and > the >>>>> e-mail >>>>>> contains patient information, please contact the Partners Compliance >>>>> HelpLine at >>>>>> http://www.partners.org/complianceline . If the e-mail was sent to > you >>>>> in error >>>>>> but does not contain patient information, please contact the sender >>> and >>>>> properly >>>>>> dispose of the e-mail. >>>>>> >>>>> >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>>> subject >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >>>> >>> >>> -- >>> Nicole Aucoin nicole at bwh.harvard.edu >>> Surgical Planning Laboratory, Brigham and Women's Hospital >>> Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 >>> http://www.slicer.org > > -- > Nicole Aucoin nicole at bwh.harvard.edu > Surgical Planning Laboratory, Brigham and Women's Hospital > Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 > http://www.slicer.org From pieper at bwh.harvard.edu Thu Apr 2 08:57:49 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 02 Apr 2009 08:57:49 -0400 Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <2d509e140903280946g221ab0f5k4bdc8da744f22fb1@mail.gmail.com> References: <200903261118.58758.simon@math.uh.edu> <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> <2d509e140903270905p7912357fr4288824ae83ec23e@mail.gmail.com> <200903271321.37239.simon@math.uh.edu> <49CD2670.6030904@bwh.harvard.edu> <2d509e140903280946g221ab0f5k4bdc8da744f22fb1@mail.gmail.com> Message-ID: <49D4B64D.9060100@bwh.harvard.edu> Hi Luca - I'll try to talk this over with Jim today and see what we can come up with. Maybe we need another file that explicitly says which of the files Plugins directory should be considered as modules. -Steve Luca Antiga wrote: > Hey, > did not know about the convert on Windows, pretty funny. > Since this happened, do you think it would be worth to change the > location of Python XML modules, e.g. to Plugins/Python or even > Modules/Python to make the system more robust? I see this could happen > again in the future, especially with local user's code. > > Luca > > 2009/3/27, Steve Pieper : >> Hi - >> >> We found which files were involved with this 'import' issue and they >> should be fixed pretty soon. >> >> In general we do need to be very careful about what ends up in the >> Plugins directory, since slicer will try to execute them with a '--xml' >> argument as part of the discovery process - so any unintended side >> effects could be very bad indeed... That's why they are in the special >> plugins directory rather than, for example, just using the PATH variable. >> >> -Steve >> >> p.s. (As Luca points out, the ImageMagick utilities are very poorly >> named - on windows there's another program named 'convert' that switches >> your disk from fat32 to ntfs! Kind of an unfriendly accident when you >> intended to convert an image to jpeg...). >> >> >> >> Simon K Alexander wrote: >>> Hi Luca, >>> >>> Changing the executable status of these files in the build directory seems >>> to clear everything up. However, the >>> originals are not exectable, so the build process is setting this somehow. >>> As you note, an executable might be >>> actually wanted though, so something will have to be changed in the build >>> to do the right thing here. >>> >>> About your 1st suggestion though: >>>> 1. making sure the python file starts with #!/usr/bin/env python >>> >>> All of these files start with the XML= " ..." string. Does anything >>> potentially rely on that? >>> >>> cheers, >>> Simon >>> >>> On Friday 27 March 2009 11:05:45 am Luca Antiga wrote: >>>> Hi guys, >>>> Steve is right, this happens when the Python files have executable >>>> permissions and there's no >>>> #!/usr/bin/env python >>>> at the beginning of the file, so they are executed using the bash >>>> interpreter, which in turn calls >>>> import >>>> which is the screenshot acquisition executable from Image Magick (I >>>> always >>>> wondered who had >>>> the idea of calling Image Magick executables 'convert', 'import', >>>> 'identify', 'display'... they should >>>> be given a fine for this!) >>>> I haven't tried myself (I'm currently at a meeting) but the problem >>>> should >>>> be fixed by >>>> 1. making sure the python file starts with #!/usr/bin/env python >>>> 2. removing executable permissions >>>> 3. eventually avoiding to try to execute .py files during module >>>> discovery, >>>> although this is a double-edged >>>> knife, since it would prevent real executables written in python and >>>> having >>>> a py extension to be discovered. >>>> >>>> Cheers! >>>> >>>> Luca >>>> >>>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> > From fedorov at bwh.harvard.edu Thu Apr 2 12:47:49 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Thu, 2 Apr 2009 12:47:49 -0400 Subject: [slicer-devel] Slice translation in slice viewer In-Reply-To: <82f536810903310613q62165ac7u35fd0818fa12ba86@mail.gmail.com> References: <82f536810903301601v4551ba74p8c35e6a1cf26a758@mail.gmail.com> <49D21172.9090508@bwh.harvard.edu> <82f536810903310613q62165ac7u35fd0818fa12ba86@mail.gmail.com> Message-ID: <82f536810904020947m5d871fdfmd6da07a42be8d3da@mail.gmail.com> The workaround has been found, thanks to Kilian, who has very similar platform. The solution is to press both left and right mouse buttons at the same time, this will drag the slice view. I will update the wiki to have this trick documented. On Tue, Mar 31, 2009 at 9:13 AM, Andriy Fedorov wrote: > Steve, > > Yes, the middle button is functional otherwise (i.e., I can paste in > an xterm). But not in Slicer -- in 3D viewer middle button does zoom, > not translation. It is interesting that I have the same behavior with > both the the external mouse: right button zoom, middle button scroll > through slices. > > Maybe it is just something specific to my system setup, but I didn't > do anything special. > > It would be nice if there was more than one way to do slice > translation, in any case. > > Andriy Fedorov > > > On Tue, Mar 31, 2009 at 8:49 AM, Steve Pieper wrote: >> Hi Andriy - >> >> We may need to add a workaround for that. ?Does the middle mouse work >> correctly elsewhere? ?I.e. can you translate the 3D viewer in slicer? Can >> you paste in an xterm? >> >> -Steve >> >> Andriy Fedorov wrote: >>> >>> Hi, >>> >>> I was told that when the middle mouse button is pressed, the slice in >>> slice viewer should translate following the mouse pointer. This >>> functionality doesn't work on my platform (reported in >>> http://na-mic.org/Mantis/view.php?id=460). >>> >>> Is there any workaround to re-center/translate slice views in Slicer? >>> I know I can create a transform or change image origin, but I would >>> rather not do that, because I need this only for visualization. >>> >>> Any hint is appreciated. >>> >>> Andriy Fedorov >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >> > From nagy.attila at yahoo.com Thu Apr 2 13:17:00 2009 From: nagy.attila at yahoo.com (nagy.attila at yahoo.com) Date: Thu, 2 Apr 2009 10:17:00 -0700 (PDT) Subject: [slicer-devel] Slice translation in slice viewer Message-ID: <692528.14811.qm@web37403.mail.mud.yahoo.com> Hey guys, on Solaris things function this way: left mouse button: (nothing special) middle scroll wheel when pressed and hold: x-y movement of the active slice in view scroll wheel "scrolled": advance between slices forward-backward (i-s, l-r, a-p) right mouse button click and hold: zoom active slice (a little different than in http://na-mic.org/Mantis/view.php?id=460) (I have to add that this is a 64bit build "patched" with the all-new-libs) Attila --- On Thu, 4/2/09, Andriy Fedorov wrote: > From: Andriy Fedorov > Subject: Re: [slicer-devel] Slice translation in slice viewer > To: "Steve Pieper" > Cc: "Slicer Devel List" > Date: Thursday, April 2, 2009, 6:47 PM > The workaround has been found, thanks > to Kilian, who has very similar platform. > > The solution is to press both left and right mouse buttons > at the same > time, this will drag the slice view. I will update the wiki > to have > this trick documented. > > On Tue, Mar 31, 2009 at 9:13 AM, Andriy Fedorov > wrote: > > Steve, > > > > Yes, the middle button is functional otherwise (i.e., > I can paste in > > an xterm). But not in Slicer -- in 3D viewer middle > button does zoom, > > not translation. It is interesting that I have the > same behavior with > > both the the external mouse: right button zoom, middle > button scroll > > through slices. > > > > Maybe it is just something specific to my system > setup, but I didn't > > do anything special. > > > > It would be nice if there was more than one way to do > slice > > translation, in any case. > > > > Andriy Fedorov > > > > > > On Tue, Mar 31, 2009 at 8:49 AM, Steve Pieper > wrote: > >> Hi Andriy - > >> > >> We may need to add a workaround for that. ?Does > the middle mouse work > >> correctly elsewhere? ?I.e. can you translate the > 3D viewer in slicer? Can > >> you paste in an xterm? > >> > >> -Steve > >> > >> Andriy Fedorov wrote: > >>> > >>> Hi, > >>> > >>> I was told that when the middle mouse button > is pressed, the slice in > >>> slice viewer should translate following the > mouse pointer. This > >>> functionality doesn't work on my platform > (reported in > >>> http://na-mic.org/Mantis/view.php?id=460). > >>> > >>> Is there any workaround to re-center/translate > slice views in Slicer? > >>> I know I can create a transform or change > image origin, but I would > >>> rather not do that, because I need this only > for visualization. > >>> > >>> Any hint is appreciated. > >>> > >>> Andriy Fedorov > >>> > _______________________________________________ > >>> slicer-devel mailing list > >>> slicer-devel at bwh.harvard.edu > >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > >>> To unsubscribe: send email to > >>> slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the > >>> subject > >> > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From luca.antiga at gmail.com Fri Apr 3 03:45:35 2009 From: luca.antiga at gmail.com (Luca Antiga) Date: Fri, 3 Apr 2009 09:45:35 +0200 Subject: [slicer-devel] sorry to triple post (again on the screenshots) In-Reply-To: <49D4B64D.9060100@bwh.harvard.edu> References: <200903261118.58758.simon@math.uh.edu> <38453.198.163.94.5.1238093079.squirrel@mail.knowledgevis.com> <2d509e140903270905p7912357fr4288824ae83ec23e@mail.gmail.com> <200903271321.37239.simon@math.uh.edu> <49CD2670.6030904@bwh.harvard.edu> <2d509e140903280946g221ab0f5k4bdc8da744f22fb1@mail.gmail.com> <49D4B64D.9060100@bwh.harvard.edu> Message-ID: <7B4C8402-685B-4BDE-9C2E-D59A36DEEA48@gmail.com> Hey Steve, great that you're looking into this. The file that explicitly states which files to load is a possibility that would also work for Modules, especially when an externally developed module becomes not "in sync" with the Slicer build and Slicer hangs during startup. The only way around it now is either to recompile your external module (good for developers but not for users) or to move the module out of the way, since one needs to be in Slicer to un-tick it from the module list in the module selection panel. One idea could be a small external application that replicates the modules panel and allows to un-tick modules prior to going through the discovery process. Or, maybe better, a "safe mode" for Slicer with which it skips the module discovery, doesn't load any module but allows one to access the module list. Another idea to make things safer somehow is that each module discovery could take place in a subprocess, which Slicer monitors for success and eventually times out if it hangs or if something weird happens, like in the screenshots case, or something like it. If everything is ok, then the plugin or module is officially loaded in the main process. The downside would be that the whole thing would almost take twice the time. Just thinking aloud, I apologize if the above doesn't make sense :-) Cheers Luca On Apr 2, 2009, at 2:57 PM, Steve Pieper wrote: > Hi Luca - > > I'll try to talk this over with Jim today and see what we can come > up with. Maybe we need another file that explicitly says which of > the files Plugins directory should be considered as modules. > > -Steve > > > Luca Antiga wrote: >> Hey, >> did not know about the convert on Windows, pretty funny. >> Since this happened, do you think it would be worth to change the >> location of Python XML modules, e.g. to Plugins/Python or even >> Modules/Python to make the system more robust? I see this could >> happen >> again in the future, especially with local user's code. >> Luca >> 2009/3/27, Steve Pieper : >>> Hi - >>> >>> We found which files were involved with this 'import' issue and they >>> should be fixed pretty soon. >>> >>> In general we do need to be very careful about what ends up in the >>> Plugins directory, since slicer will try to execute them with a '-- >>> xml' >>> argument as part of the discovery process - so any unintended side >>> effects could be very bad indeed... That's why they are in the >>> special >>> plugins directory rather than, for example, just using the PATH >>> variable. >>> >>> -Steve >>> >>> p.s. (As Luca points out, the ImageMagick utilities are very poorly >>> named - on windows there's another program named 'convert' that >>> switches >>> your disk from fat32 to ntfs! Kind of an unfriendly accident when >>> you >>> intended to convert an image to jpeg...). >>> >>> >>> >>> Simon K Alexander wrote: >>>> Hi Luca, >>>> >>>> Changing the executable status of these files in the build >>>> directory seems >>>> to clear everything up. However, the >>>> originals are not exectable, so the build process is setting this >>>> somehow. >>>> As you note, an executable might be >>>> actually wanted though, so something will have to be changed in >>>> the build >>>> to do the right thing here. >>>> >>>> About your 1st suggestion though: >>>>> 1. making sure the python file starts with #!/usr/bin/env python >>>> >>>> All of these files start with the XML= " ..." string. Does >>>> anything >>>> potentially rely on that? >>>> >>>> cheers, >>>> Simon >>>> >>>> On Friday 27 March 2009 11:05:45 am Luca Antiga wrote: >>>>> Hi guys, >>>>> Steve is right, this happens when the Python files have executable >>>>> permissions and there's no >>>>> #!/usr/bin/env python >>>>> at the beginning of the file, so they are executed using the bash >>>>> interpreter, which in turn calls >>>>> import >>>>> which is the screenshot acquisition executable from Image Magick >>>>> (I >>>>> always >>>>> wondered who had >>>>> the idea of calling Image Magick executables 'convert', 'import', >>>>> 'identify', 'display'... they should >>>>> be given a fine for this!) >>>>> I haven't tried myself (I'm currently at a meeting) but the >>>>> problem >>>>> should >>>>> be fixed by >>>>> 1. making sure the python file starts with #!/usr/bin/env python >>>>> 2. removing executable permissions >>>>> 3. eventually avoiding to try to execute .py files during module >>>>> discovery, >>>>> although this is a double-edged >>>>> knife, since it would prevent real executables written in python >>>>> and >>>>> having >>>>> a py extension to be discovered. >>>>> >>>>> Cheers! >>>>> >>>>> Luca >>>>> >>>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as >>>> the >>>> subject >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject >>> -- Luca Antiga, PhD Head, Medical Imaging Unit, Biomedical Engineering Department, Mario Negri Institute. mail: Villa Camozzi, 24020, Ranica (BG), Italy phone: +39 035 4535-381 email: antiga at marionegri.it web: http://villacamozzi.marionegri.it/~luca -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090403/229b8dc4/attachment-0001.html From vrnova at gmail.com Fri Apr 3 13:22:14 2009 From: vrnova at gmail.com (Yanling Liu) Date: Fri, 3 Apr 2009 13:22:14 -0400 Subject: [slicer-devel] itkDicomImageIO2Factory.h: No such file or directory Message-ID: <24ceacd90904031022p6ba1df2fm615c7f9418448879@mail.gmail.com> Hello...I got this error message (see title) after a fresh rebuild... Details: http://www.cdash.org/CDash/viewBuildError.php?buildid=305443 Any help? Thanks! Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090403/4f574521/attachment.html From pieper at bwh.harvard.edu Fri Apr 3 13:56:24 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 03 Apr 2009 13:56:24 -0400 Subject: [slicer-devel] itkDicomImageIO2Factory.h: No such file or directory In-Reply-To: <24ceacd90904031022p6ba1df2fm615c7f9418448879@mail.gmail.com> References: <24ceacd90904031022p6ba1df2fm615c7f9418448879@mail.gmail.com> Message-ID: <49D64DC8.3040003@bwh.harvard.edu> Hi Yanling - Sorry about that - I checked in a fix a few minutes ago. (the filename needed to be itkDICOMImageIO2Factory.h but windows ignored the capitalization so I didn't see this right away...) -Steve Yanling Liu wrote: > Hello...I got this error message (see title) after a fresh rebuild... > > Details: http://www.cdash.org/CDash/viewBuildError.php?buildid=305443 > > Any help? Thanks! > Yanling > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From vrnova at gmail.com Fri Apr 3 14:29:08 2009 From: vrnova at gmail.com (Yanling Liu) Date: Fri, 3 Apr 2009 14:29:08 -0400 Subject: [slicer-devel] not able to change window layout Message-ID: <24ceacd90904031129v71c30016k3bcaa8a668cd2027@mail.gmail.com> Hi, after a fresh rebuild I couldn't change Slicer window layout anymore. Does anyone else see this problem too? Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090403/4014c83f/attachment.html From fedorov at bwh.harvard.edu Fri Apr 3 14:53:03 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Fri, 3 Apr 2009 14:53:03 -0400 Subject: [slicer-devel] not able to change window layout In-Reply-To: <24ceacd90904031129v71c30016k3bcaa8a668cd2027@mail.gmail.com> References: <24ceacd90904031129v71c30016k3bcaa8a668cd2027@mail.gmail.com> Message-ID: <82f536810904031153hda4f2a9t4537818d6a6747c6@mail.gmail.com> I didn't update my local Slicer source, but the latest nightly build has the same problem for me as you describe. In addition to this, It appears that none of the menu items corresponding to the "Load data" icon work (no response). On Fri, Apr 3, 2009 at 2:29 PM, Yanling Liu wrote: > Hi, after a fresh rebuild I couldn't change Slicer window layout anymore. > Does anyone else see this problem too? > > Thanks, > Yanling > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > From wjp at bwh.harvard.edu Fri Apr 3 14:59:31 2009 From: wjp at bwh.harvard.edu (Wendy Plesniak) Date: Fri, 03 Apr 2009 14:59:31 -0400 Subject: [slicer-devel] not able to change window layout In-Reply-To: <24ceacd90904031129v71c30016k3bcaa8a668cd2027@mail.gmail.com> References: <24ceacd90904031129v71c30016k3bcaa8a668cd2027@mail.gmail.com> Message-ID: <49D65C93.3010502@bwh.harvard.edu> Hi Yanling, I think this is fixed now, if you update in Slicer/Base/GUI. Let me know if you have more trouble. -wen Yanling Liu wrote: > Hi, after a fresh rebuild I couldn't change Slicer window layout > anymore. Does anyone else see this problem too? > > Thanks, > Yanling > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From vrnova at gmail.com Fri Apr 3 15:29:49 2009 From: vrnova at gmail.com (Yanling Liu) Date: Fri, 3 Apr 2009 15:29:49 -0400 Subject: [slicer-devel] screen capture Message-ID: <24ceacd90904031229h3957c923ra882dad63fd1ff56@mail.gmail.com> Hi, the new screen capture feature is great! One small suggestion: is it possible to minimize or hide the "Screen Capture Options" dialog window while capturing Slicer rendering window? This could be useful when user are trying to capture full screen images. Thanks, Yanling -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090403/a0d5f7ae/attachment.html From inorton at partners.org Fri Apr 3 16:31:03 2009 From: inorton at partners.org (Isaiah Norton) Date: Fri, 03 Apr 2009 16:31:03 -0400 Subject: [slicer-devel] screen capture In-Reply-To: <24ceacd90904031229h3957c923ra882dad63fd1ff56@mail.gmail.com> References: <24ceacd90904031229h3957c923ra882dad63fd1ff56@mail.gmail.com> Message-ID: <1238790663.27731.31.camel@thanks.bwh.harvard.edu> Thanks for adding this, looks good and will be helpful!! It seems to momentarily flip the scene around when I click capture, then snaps back. But the actual images look fine. One other small suggestion. The cap. window is modal right now. Would it be possible to make this less stringent? ie force the capture window to stickyfloat on top of the main window, but don't block the mouse. This would make capture relatively more seamless (-2 mouseclicks!) Alternative if supported by kwwidget might be click-hold on the button to get the options window, quick-click to capture once configured. Or a keyboard shortcut. -Isaiah On Fri, 2009-04-03 at 15:29 -0400, Yanling Liu wrote: > Hi, the new screen capture feature is great! One small suggestion: is > it possible to minimize or hide the "Screen Capture Options" dialog > window while capturing Slicer rendering window? This could be useful > when user are trying to capture full screen images. > > Thanks, > Yanling > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From harish.slicer at gmail.com Sun Apr 5 20:54:58 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Sun, 5 Apr 2009 17:54:58 -0700 Subject: [slicer-devel] Testing Command Line Interface Message-ID: Hi all, I am new to the developing environment in slicer. I am learning slicer3 architecture from the tutorials present on the web. I understood how to integrate our own modules as CLI in Slicer. But I have one question regarding testing. I have build Slicer in Microsoft Visual Studio 2008 with a CLI module. I got the Slicer.sln file and the slicer executable. Now I made some changes to the .cxx file (in the slicer.sln) for that module and re-build that module. I have the executables in SLICER_HOME/slicer-build/lib/slicer3/plugins/debug directory. How should I test this executable. I mean if I have an executable .exe, how could I give input to it and test it. Is there any script which does that. I just want to test only that module (by printing some printf statements etc) I would be very grateful if someone could help me. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090405/f7ba3f6d/attachment.html From nagy.attila at yahoo.com Mon Apr 6 06:28:47 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Mon, 6 Apr 2009 03:28:47 -0700 (PDT) Subject: [slicer-devel] Scene save/load bug? Message-ID: <641692.44020.qm@web37405.mail.mud.yahoo.com> Hi all, is this normal? www.orl.szote.u-szeged.hu/~attila/scene-reload-bug.png After saving a scene, and loading it back, the ct slices are rotated, but labelmaps aren't. (I assume, nothing should be rotated... :)) Is this a known problem, or should I update my build? (it's from the 24 March) Of course, it's Solaris :), full 64 bit build. Attila From pieper at bwh.harvard.edu Mon Apr 6 08:07:25 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 06 Apr 2009 08:07:25 -0400 Subject: [slicer-devel] new batchmake and openigtlink Message-ID: <49D9F07D.3010408@bwh.harvard.edu> FYI: associated with some bug fixes the trunk build now needs updated versions of OpenIGTLink and BatchMake. Doing something like the following from the source dir should work for most people: rm ../Slicer3-lib/OpenIGTLink* rm ../Slicer3-lib/BatchMake* ./Scripts/getbuildtest.tcl -Steve From fedorov at bwh.harvard.edu Mon Apr 6 13:30:57 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 6 Apr 2009 13:30:57 -0400 Subject: [slicer-devel] Testing Command Line Interface In-Reply-To: References: Message-ID: <82f536810904061030g709e7fa8k4b812855f219d77f@mail.gmail.com> Harish, You should be able to run your module by doing something like this: \Slicer3 --launch This should print out "help" of your module. I am using Linux, but this should work on Windows as well. Please update the list if this doesn't work for you. Andriy Fedorov On Sun, Apr 5, 2009 at 8:54 PM, Harish Doddi wrote: > > > Hi all, > > I am new to the developing environment in slicer. I am learning slicer3 > architecture from the tutorials present on the web. I understood how to > integrate our own modules as CLI in Slicer. But I have one question > regarding testing. I have build Slicer in Microsoft Visual Studio 2008 with > a CLI module. I got the Slicer.sln file and the slicer executable. Now I > made some changes to the .cxx file (in the slicer.sln) for that module and > re-build that module. I have the executables in > SLICER_HOME/slicer-build/lib/slicer3/plugins/debug directory. How should I > test this executable. I mean if I have an executable .exe, how > could I give input to it and test it. Is there any script which does that. I > just want to test only that module (by printing some printf statements etc) > > > I would be very grateful if someone could help me. > > Thanks, > Harish > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > From harish.slicer at gmail.com Mon Apr 6 13:46:29 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 6 Apr 2009 10:46:29 -0700 Subject: [slicer-devel] Testing Command Line Interface In-Reply-To: <82f536810904061030g709e7fa8k4b812855f219d77f@mail.gmail.com> References: <82f536810904061030g709e7fa8k4b812855f219d77f@mail.gmail.com> Message-ID: Hi Andriy, Thanks for your email. I tried out the following command in the Slicer3-build directory ./Slicer3 --launch But it doesnt print anything (though I have some printf statements in my .cxx file) . The prompt comes back after hitting this command. Am I missing something ? Thanks, Harish On Mon, Apr 6, 2009 at 10:30 AM, Andriy Fedorov wrote: > Harish, > > You should be able to run your module by doing something like this: > > \Slicer3 --launch > > This should print out "help" of your module. I am using Linux, but > this should work on Windows as well. Please update the list if this > doesn't work for you. > > Andriy Fedorov > > > On Sun, Apr 5, 2009 at 8:54 PM, Harish Doddi > wrote: > > > > > > Hi all, > > > > I am new to the developing environment in slicer. I am learning slicer3 > > architecture from the tutorials present on the web. I understood how to > > integrate our own modules as CLI in Slicer. But I have one question > > regarding testing. I have build Slicer in Microsoft Visual Studio 2008 > with > > a CLI module. I got the Slicer.sln file and the slicer executable. Now I > > made some changes to the .cxx file (in the slicer.sln) for that module > and > > re-build that module. I have the executables in > > SLICER_HOME/slicer-build/lib/slicer3/plugins/debug directory. How should > I > > test this executable. I mean if I have an executable .exe, > how > > could I give input to it and test it. Is there any script which does > that. I > > just want to test only that module (by printing some printf statements > etc) > > > > > > I would be very grateful if someone could help me. > > > > Thanks, > > Harish > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu > > with unsubscribe as the subject > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090406/479decf3/attachment.html From fedorov at bwh.harvard.edu Mon Apr 6 14:35:04 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 6 Apr 2009 14:35:04 -0400 Subject: [slicer-devel] Testing Command Line Interface In-Reply-To: References: <82f536810904061030g709e7fa8k4b812855f219d77f@mail.gmail.com> Message-ID: <82f536810904061135u74a4bfap9123ff00c5ac7ee5@mail.gmail.com> Harish, Yes, this is actually "normal" behavior on Windows. I was not aware of this until Steve explained. The problem is that Slicer is compiled as a GUI application, and redirection of stdout and stderr streams is not working the same way as it does on Linux. A workaround that appears to solve the stream redirection problem is to use use rxvt terminal (you need to add this package in your Cygwin installation). If you use rxvt, you will be able to see the output of your printf()'s on the console, in case you still want to use this for debugging. Hope this helps... Andriy Fedorov On Mon, Apr 6, 2009 at 1:46 PM, Harish Doddi wrote: > Hi Andriy, > > Thanks for your email. I tried out the following command in the > Slicer3-build directory > > ./Slicer3 --launch > > But it doesnt print anything (though I have some printf statements in my > .cxx file) . The prompt comes back after hitting this command.? Am I missing > something ? > > > Thanks, > Harish > > > > On Mon, Apr 6, 2009 at 10:30 AM, Andriy Fedorov > wrote: >> >> Harish, >> >> You should be able to run your module by doing something like this: >> >> \Slicer3 --launch >> >> This should print out "help" of your module. I am using Linux, but >> this should work on Windows as well. Please update the list if this >> doesn't work for you. >> >> Andriy Fedorov >> >> >> On Sun, Apr 5, 2009 at 8:54 PM, Harish Doddi >> wrote: >> > >> > >> > Hi all, >> > >> > I am new to the developing environment in slicer. I am learning slicer3 >> > architecture from the tutorials present on the web. I understood how to >> > integrate our own modules as CLI in Slicer. But I have one question >> > regarding testing. I have build Slicer in Microsoft Visual Studio 2008 >> > with >> > a CLI module. I got the Slicer.sln file and the slicer executable. Now I >> > made some changes to the .cxx file (in the slicer.sln) for that module >> > and >> > re-build that module. I have the executables in >> > SLICER_HOME/slicer-build/lib/slicer3/plugins/debug directory. How should >> > I >> > test this executable. I mean if I have an executable .exe, >> > how >> > could I give input to it and test it. Is there any script which does >> > that. I >> > just want to test only that module (by printing some printf statements >> > etc) >> > >> > >> > I would be very grateful if someone could help me. >> > >> > Thanks, >> > Harish >> > >> > _______________________________________________ >> > slicer-devel mailing list >> > slicer-devel at bwh.harvard.edu >> > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> > To unsubscribe: send email to >> > slicer-devel-request at massmail.spl.harvard.edu >> > with unsubscribe as the subject >> > > > From harish.slicer at gmail.com Mon Apr 6 15:15:15 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 6 Apr 2009 12:15:15 -0700 Subject: [slicer-devel] [slicer-users] Testing Command Line Interface In-Reply-To: <49DA410B.70103@bwh.harvard.edu> References: <49DA410B.70103@bwh.harvard.edu> Message-ID: Hi Steve, I tried the thing that you mentioned. Now when I run the module with Right click on project -> Debug -> Start new instance I see the attached error message on the screen. I understand that the vtkIO.dll is present in Slicer-lib/VTK-build/bin/Debug directory. However, I have this directory included in the project properties -> C/C++ -> General -> Additional Include Directories. Is this not the place where we set up the path for searching ? Thanks, Harish On Mon, Apr 6, 2009 at 10:51 AM, Steve Pieper wrote: > Hi Harish - > > Typically what you want to do is launch visual studio in a shell that has > the correct environment (path variables) to find the > slicer/vtk/itk/teem/etc/ shared libraries. You can use slicer's built-in > launcher for this with something like one of the lines shown here: > > > http://www.na-mic.org/Wiki/index.php/User:Pieper#Launching_visual_studio_for_slicer3_debugging > > Then in visual studio you can pick your command line module from the > solution and right click to bring up context menu. Pick Set As Startup > Project to debug that program. Then you can edit the properties for the > project and add command line arguments in the debugging dialog. > > Once you've done that you can single step through using the debugger. > > -Steve > > > Harish Doddi wrote: > >> >> >> >> Hi all, >> >> I am new to the developing environment in slicer. I am learning slicer3 >> architecture from the tutorials present on the web. I understood how to >> integrate our own modules as CLI in Slicer. But I have one question >> regarding testing. I have build Slicer in Microsoft Visual Studio 2008 with >> a CLI module. I got the Slicer.sln file and the slicer executable. Now I >> made some changes to the .cxx file (in the slicer.sln) for that module and >> re-build that module. I have the executables in >> SLICER_HOME/slicer-build/lib/slicer3/plugins/debug directory. How should I >> test this executable. I mean if I have an executable .exe, how >> could I give input to it and test it. Is there any script which does that. I >> just want to test only that module (by printing some printf statements etc) >> >> >> I would be very grateful if someone could help me. >> >> Thanks, >> Harish >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-users mailing list >> slicer-users at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >> To unsubscribe: send email to >> slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090406/3f672aa5/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: Runtime_Error.JPG Type: image/jpeg Size: 258971 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090406/3f672aa5/attachment-0001.jpe From pieper at bwh.harvard.edu Mon Apr 6 15:55:36 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 06 Apr 2009 15:55:36 -0400 Subject: [slicer-devel] [slicer-users] Testing Command Line Interface In-Reply-To: References: <49DA410B.70103@bwh.harvard.edu> Message-ID: <49DA5E38.4090408@bwh.harvard.edu> Hi Harish - You shouldn't need to add any directories through the visual studio interface (the running devenv through the slicer launcher should do all that for you). To test that the launcher is working right for you, you could run this command: ./Slicer3 --launch rxvt and in the newly launched window check that the PATH variable includes your Slicer3-lib/VTK-build/bin/Debug along with the other lib directories. (rxvt is the shell that Andriy mentioned - you get it via cygwin). Hope that helps, Steve Harish Doddi wrote: > Hi Steve, > > I tried the thing that you mentioned. Now when I run the module with > Right click on project -> Debug -> Start new instance I see the attached > error message on the screen. I understand that the vtkIO.dll is present > in Slicer-lib/VTK-build/bin/Debug directory. However, I have this > directory included in the project properties -> C/C++ -> General -> > Additional Include Directories. Is this not the place where we set up > the path for searching ? > > Thanks, > Harish > > On Mon, Apr 6, 2009 at 10:51 AM, Steve Pieper > wrote: > > Hi Harish - > > Typically what you want to do is launch visual studio in a shell > that has the correct environment (path variables) to find the > slicer/vtk/itk/teem/etc/ shared libraries. You can use slicer's > built-in launcher for this with something like one of the lines > shown here: > > http://www.na-mic.org/Wiki/index.php/User:Pieper#Launching_visual_studio_for_slicer3_debugging > > Then in visual studio you can pick your command line module from the > solution and right click to bring up context menu. Pick Set As > Startup Project to debug that program. Then you can edit the > properties for the project and add command line arguments in the > debugging dialog. > > Once you've done that you can single step through using the debugger. > > -Steve > > > Harish Doddi wrote: > > > > > Hi all, > > I am new to the developing environment in slicer. I am learning > slicer3 architecture from the tutorials present on the web. I > understood how to integrate our own modules as CLI in Slicer. > But I have one question regarding testing. I have build Slicer > in Microsoft Visual Studio 2008 with a CLI module. I got the > Slicer.sln file and the slicer executable. Now I made some > changes to the .cxx file (in the slicer.sln) for that module and > re-build that module. I have the executables in > SLICER_HOME/slicer-build/lib/slicer3/plugins/debug directory. > How should I test this executable. I mean if I have an > executable .exe, how could I give input to it and > test it. Is there any script which does that. I just want to > test only that module (by printing some printf statements etc) > > > I would be very grateful if someone could help me. > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to > slicer-users-request at massmail.spl.harvard.edu > with > unsubscribe as the subject > > > > ------------------------------------------------------------------------ > From pieper at bwh.harvard.edu Mon Apr 6 16:09:50 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 06 Apr 2009 16:09:50 -0400 Subject: [slicer-devel] Scene save/load bug? In-Reply-To: <641692.44020.qm@web37405.mail.mud.yahoo.com> References: <641692.44020.qm@web37405.mail.mud.yahoo.com> Message-ID: <49DA618E.9080807@bwh.harvard.edu> Ouch, that certainly looks like a bug - I haven't seen it myself and can't recreate it. Is it something where you could create a zip file of the directory with data and mrml file to attach to a bug report? Perhaps with just nrrd files (No patient confidential information included, of course!). Thanks, Steve Attila Nagy wrote: > Hi all, > > is this normal? > > www.orl.szote.u-szeged.hu/~attila/scene-reload-bug.png > > After saving a scene, and loading it back, the ct slices are rotated, but labelmaps aren't. (I assume, nothing should be rotated... :)) > Is this a known problem, or should I update my build? (it's from the 24 March) > Of course, it's Solaris :), full 64 bit build. > > Attila > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From harish.slicer at gmail.com Mon Apr 6 20:06:02 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 6 Apr 2009 17:06:02 -0700 Subject: [slicer-devel] [slicer-users] Testing Command Line Interface In-Reply-To: <49DA5E38.4090408@bwh.harvard.edu> References: <49DA410B.70103@bwh.harvard.edu> <49DA5E38.4090408@bwh.harvard.edu> Message-ID: Hi Steve, Thanks a lot. It works fine now. Thanks, Harish On Mon, Apr 6, 2009 at 12:55 PM, Steve Pieper wrote: > Hi Harish - > > You shouldn't need to add any directories through the visual studio > interface (the running devenv through the slicer launcher should do all that > for you). To test that the launcher is working right for you, you could run > this command: > > ./Slicer3 --launch rxvt > > and in the newly launched window check that the PATH variable includes your > Slicer3-lib/VTK-build/bin/Debug along with the other lib directories. (rxvt > is the shell that Andriy mentioned - you get it via cygwin). > > Hope that helps, > Steve > > > Harish Doddi wrote: > >> Hi Steve, >> >> I tried the thing that you mentioned. Now when I run the module with Right >> click on project -> Debug -> Start new instance I see the attached error >> message on the screen. I understand that the vtkIO.dll is present in >> Slicer-lib/VTK-build/bin/Debug directory. However, I have this directory >> included in the project properties -> C/C++ -> General -> Additional Include >> Directories. Is this not the place where we set up the path for searching ? >> >> Thanks, >> Harish >> >> On Mon, Apr 6, 2009 at 10:51 AM, Steve Pieper > pieper at bwh.harvard.edu>> wrote: >> >> Hi Harish - >> >> Typically what you want to do is launch visual studio in a shell >> that has the correct environment (path variables) to find the >> slicer/vtk/itk/teem/etc/ shared libraries. You can use slicer's >> built-in launcher for this with something like one of the lines >> shown here: >> >> >> http://www.na-mic.org/Wiki/index.php/User:Pieper#Launching_visual_studio_for_slicer3_debugging >> >> Then in visual studio you can pick your command line module from the >> solution and right click to bring up context menu. Pick Set As >> Startup Project to debug that program. Then you can edit the >> properties for the project and add command line arguments in the >> debugging dialog. >> >> Once you've done that you can single step through using the debugger. >> >> -Steve >> >> >> Harish Doddi wrote: >> >> >> >> >> Hi all, >> >> I am new to the developing environment in slicer. I am learning >> slicer3 architecture from the tutorials present on the web. I >> understood how to integrate our own modules as CLI in Slicer. >> But I have one question regarding testing. I have build Slicer >> in Microsoft Visual Studio 2008 with a CLI module. I got the >> Slicer.sln file and the slicer executable. Now I made some >> changes to the .cxx file (in the slicer.sln) for that module and >> re-build that module. I have the executables in >> SLICER_HOME/slicer-build/lib/slicer3/plugins/debug directory. >> How should I test this executable. I mean if I have an >> executable .exe, how could I give input to it and >> test it. Is there any script which does that. I just want to >> test only that module (by printing some printf statements etc) >> >> >> I would be very grateful if someone could help me. >> >> Thanks, >> Harish >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-users mailing list >> slicer-users at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >> To unsubscribe: send email to >> slicer-users-request at massmail.spl.harvard.edu >> with >> unsubscribe as the subject >> >> >> >> ------------------------------------------------------------------------ >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090406/898de173/attachment.html From harish.slicer at gmail.com Mon Apr 6 21:45:55 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 6 Apr 2009 18:45:55 -0700 Subject: [slicer-devel] Cannot read label map Message-ID: Hi all, I am testing a module called "PolyDataFilledToLabelMap" through command line interface. I run the following command in the Slicer-build/lib/S ./PolyDataToFilledMap.exe --help and the output is USAGE: c:\Documents and Settings\harishd\Desktop\Nicole Aucoin\slicer_source_code_4\Slicer3 -build\lib\Slicer3\Plugins\Debug\Po lyDataToFilledLabelMap.exe [--processinformationaddress ] [--xml] [--echo] [-s >] [-l ] [-i ] [-m ] [--] [--version] [-h] Where: --processinformationaddress Address of a structure to store process information (progress, abort, etc.). (default: 0) --xml Produce xml description of command line arguments (default: 0) --echo Echo the command line arguments (default: 0) -s >, --seed > Seed point for flood fill, use model center of mass (default: 0,0,0) -l , --outputVolume The output label volume -i , --labelVolume Input label volume to start from -m , --model Model --, --ignore_rest Ignores the rest of the labeled arguments following this flag. --version Displays version information and exits. -h, --help Displays usage information and exits. Description: Intersects an input model with an input volume and produces an output label map. I run the module by supplying the following command ./PolyDataToFilledLabelMap.exe -s 6.41548,10.6075,29.8513 -l /cygdrive/c/Documents\ and\ Settings/harishd/Desktop/Nicole\ Aucoin/slicer_source_code_4/Scene/New_Scene_Femur/Label_Map_58_out.nrrd -i /cygdrive/c/Documents\ and\ Settings/harishd/Desktop/Nicole\ Aucoin/slicer_source_code_4/Scene/New_Scene_Femur/Label_Map_58.nrrd -m /cygdrive/c/Documents\ and\ Settings/harishd/Desktop/Nicole\ Aucoin/slicer_source_code_4/Scene/New_Scene_Femur/58_femur_right_center_closed.vtk I get the following error message Couldn't open file: /cygdrive/c/Documents and Settings/harishd/Desktop/Nicole Aucoin/slicer_source_code_4/Scene/New_Scene_Femur/Label_Map_58.nrrd c:\Documents and Settings\harishd\Desktop\Nicole Aucoin\slicer_source_code_4\Slicer3-build\lib\Slicer3\Plugins\Debug\PolyDataToFilledLabelMap.exe: exception caught ! itk::ImageFileReaderException (013CE618) Location: "void __thiscall itk::ImageFileReader,class itk::DefaultConvertPixelTraits >::GenerateOutputInformation(void)" File: c:\documents and settings\harishd\desktop\nicole aucoin\slicer_source_code_4\slicer3-lib\insight\code\io\itkImageFileReader.txx Line: 144 Description: Could not create IO object for file /cygdrive/c/Documents and Settings/harishd/Desktop/Nicole Aucoin/slicer_source_code_4/Scene/New_Scene_Femur/Label_Map_58.nrrd The file doesn't exist. Filename = /cygdrive/c/Documents and Settings/harishd/Desktop/Nicole Aucoin/slicer_source_code_4/Scene/New_Scene_Femur/Label_Map_58.nrrd However, I have the the file Label_Map_58.nrrd in the appropriate directory. I do not understand why it says "File doesnt exist" Could somebody help me on this ? Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090406/acd4a641/attachment-0001.html From pieper at bwh.harvard.edu Tue Apr 7 18:09:18 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 07 Apr 2009 18:09:18 -0400 Subject: [slicer-devel] update on bugs and release timing Message-ID: <49DBCF0E.6060606@bwh.harvard.edu> Hi - After going through the bug list we've identified 19 bugs that are tagged as "3.4 Targeted fix". The link should get you to them or you can see the attached image. http://tinyurl.com/cu9bzp Our plan now is to create the 3.4 release branch on Tuesday April 14. ** Any fixes after that date will need to be manually checked into both the trunk and the release branch. Please keep fixing bugs and testing the nightly builds. Add any new issues to the bug tracker and flag them if you think they are critical for the release. Thanks, Steve -------------- next part -------------- A non-text attachment was scrubbed... Name: 3.4 tagged bugs.png Type: image/png Size: 218387 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090407/bd7c6a90/attachment-0001.png From harish.slicer at gmail.com Wed Apr 8 13:36:21 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 8 Apr 2009 10:36:21 -0700 Subject: [slicer-devel] Resample Volume 2 Module Message-ID: Hi all, There used to be a module called "Resample Volume 2" in the old version of slicer. However in the new version I see the following modules Resample Scalar Volume , Resample DTI Volume, Resample Scalar/Vector/DTI volume. Which module corresponds to Resample volume 2 ? Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090408/058f5052/attachment.html From bill.lorensen at gmail.com Wed Apr 8 13:59:33 2009 From: bill.lorensen at gmail.com (Bill Lorensen) Date: Wed, 8 Apr 2009 13:59:33 -0400 Subject: [slicer-devel] Resample Volume 2 Module In-Reply-To: References: Message-ID: <4db4735c0904081059l253d068aq9d7b9a66f52665fd@mail.gmail.com> I believe it is: Resample Scalar/Vector/DTI volume. The old name was non informative. On Wed, Apr 8, 2009 at 1:36 PM, Harish Doddi wrote: > > > Hi all, > > There used to be a module called "Resample Volume 2" in the old version of > slicer. However in the new version I see the following modules Resample > Scalar Volume , Resample DTI Volume, Resample Scalar/Vector/DTI volume. > Which module corresponds to Resample volume 2 ? > > > Thanks, > Harish > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > From fedorov at bwh.harvard.edu Wed Apr 8 19:45:56 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Wed, 8 Apr 2009 19:45:56 -0400 Subject: [slicer-devel] Saving DICOM header Message-ID: <82f536810904081645n45133ad4wd2a6325bb0f26c38@mail.gmail.com> Hi, I am curious if there is a tool to extract and save (parts of) DICOM header? I see Slicer is able to display DICOM header, but I would like to store some of these (like acquisition parameters, patient history) in a text file. I assume I can find the DICOM parser part in Slicer, and make a separate tool, but maybe something like this is already available. Thanks Andriy Fedorov From INORTON at PARTNERS.ORG Wed Apr 8 20:40:00 2009 From: INORTON at PARTNERS.ORG (Norton, Isaiah) Date: Wed, 8 Apr 2009 20:40:00 -0400 Subject: [slicer-devel] Saving DICOM header References: <82f536810904081645n45133ad4wd2a6325bb0f26c38@mail.gmail.com> Message-ID: <30A5D5C0F8C31B46AB1B8A7A3E18F93A0322BFC7@PHSXMB6.partners.org> I usually use "dicom2 -t1" from Sebastien Barre @ http://barre.nom.fr/medical/dicom2/ There is also dcdump in GDCM and dcmdump in DCMTK. I'm pretty sure there is a python-dicom library but I don't know the link off hand. I think the Slicer implementation is Slicer ->ITK ->GDCM. Hope that helps. -I -----Original Message----- From: slicer-devel-bounces at bwh.harvard.edu on behalf of Andriy Fedorov Sent: Wed 4/8/2009 7:45 PM To: Slicer Devel List Subject: [slicer-devel] Saving DICOM header Hi, I am curious if there is a tool to extract and save (parts of) DICOM header? I see Slicer is able to display DICOM header, but I would like to store some of these (like acquisition parameters, patient history) in a text file. I assume I can find the DICOM parser part in Slicer, and make a separate tool, but maybe something like this is already available. Thanks Andriy Fedorov _______________________________________________ slicer-devel mailing list slicer-devel at bwh.harvard.edu http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From fedorov at bwh.harvard.edu Wed Apr 8 22:40:18 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Wed, 8 Apr 2009 22:40:18 -0400 Subject: [slicer-devel] Saving DICOM header In-Reply-To: <30A5D5C0F8C31B46AB1B8A7A3E18F93A0322BFC7@PHSXMB6.partners.org> References: <82f536810904081645n45133ad4wd2a6325bb0f26c38@mail.gmail.com> <30A5D5C0F8C31B46AB1B8A7A3E18F93A0322BFC7@PHSXMB6.partners.org> Message-ID: <82f536810904081940t5428ed66x9369600dc3c9b374@mail.gmail.com> Isaiah, Thanks for the pointer -- I think this is exactly what I was looking for. AF On Wed, Apr 8, 2009 at 8:40 PM, Norton, Isaiah wrote: > I usually use "dicom2 -t1" from Sebastien Barre @ > http://barre.nom.fr/medical/dicom2/ > > There is also dcdump in GDCM and dcmdump in DCMTK. I'm pretty sure there is a > python-dicom library but I don't know the link off hand. I think the Slicer > implementation is Slicer ->ITK ->GDCM. > > Hope that helps. > > -I > > > -----Original Message----- > From: slicer-devel-bounces at bwh.harvard.edu on behalf of Andriy Fedorov > Sent: Wed 4/8/2009 7:45 PM > To: Slicer Devel List > Subject: [slicer-devel] Saving DICOM header > > Hi, > > I am curious if there is a tool to extract and save (parts of) DICOM > header? I see Slicer is able to display DICOM header, but I would like > to store some of these (like acquisition parameters, patient history) > in a text file. > > I assume I can find the DICOM parser part in Slicer, and make a > separate tool, but maybe something like this is already available. > > Thanks > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with > unsubscribe as the subject > > From hans-johnson at uiowa.edu Thu Apr 9 15:12:18 2009 From: hans-johnson at uiowa.edu (Hans Johnson) Date: Thu, 09 Apr 2009 14:12:18 -0500 Subject: [slicer-devel] Failed build Message-ID: I am getting a failure when trying to build Slicer3: ./Scripts/getbuildtest.tcl --release -u running: svn co http://svn.scipy.org/svn/numpy/branches/1.2.x numpy svn: 'numpy' is already a working copy for a different URL Hans Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged. If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited. Please reply to the sender that you have received the message in error, then delete it. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090409/f2911dc9/attachment.html From pieper at bwh.harvard.edu Thu Apr 9 16:00:08 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 09 Apr 2009 16:00:08 -0400 Subject: [slicer-devel] Failed build In-Reply-To: References: Message-ID: <49DE53C8.2080306@bwh.harvard.edu> Hi Hans - For now you can just rm -rf ../Slicer3/python* and rerrun getbuildtest. I guess we need to catch that case - it's been coming up for BatchMake and OpenIGTLink also when the versions get updated. -Steve Hans Johnson wrote: > I am getting a failure when trying to build Slicer3: > > ./Scripts/getbuildtest.tcl --release -u > > running: svn co http://svn.scipy.org/svn/numpy/branches/1.2.x numpy > svn: 'numpy' is already a working copy for a different URL > > Hans > > Notice: This UI Health Care e-mail (including attachments) is covered by > the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is > confidential and may be legally privileged. If you are not the intended > recipient, you are hereby notified that any retention, dissemination, > distribution, or copying of this communication is strictly prohibited. > Please reply to the sender that you have received the message in error, > then delete it. Thank you. > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From millerjv at crd.ge.com Mon Apr 13 08:32:39 2009 From: millerjv at crd.ge.com (Miller, James V (GE, Research)) Date: Mon, 13 Apr 2009 08:32:39 -0400 Subject: [slicer-devel] Saving DICOM header In-Reply-To: <82f536810904081645n45133ad4wd2a6325bb0f26c38@mail.gmail.com> References: <82f536810904081645n45133ad4wd2a6325bb0f26c38@mail.gmail.com> Message-ID: <81B5B66B035911478BFFDF078E0D810205176854@CINMLVEM14.e2k.ad.ge.com> Hey Andriy, I usually use the tools from dcmtk, in particular dcmdump, to dump out DICOM headers. Jim Miller Senior Scientist GE Research Visualization & Computer Vision GE imagination at work -----Original Message----- From: slicer-devel-bounces at bwh.harvard.edu [mailto:slicer-devel-bounces at bwh.harvard.edu] On Behalf Of Andriy Fedorov Sent: Wednesday, April 08, 2009 7:46 PM To: Slicer Devel List Subject: [slicer-devel] Saving DICOM header Hi, I am curious if there is a tool to extract and save (parts of) DICOM header? I see Slicer is able to display DICOM header, but I would like to store some of these (like acquisition parameters, patient history) in a text file. I assume I can find the DICOM parser part in Slicer, and make a separate tool, but maybe something like this is already available. Thanks Andriy Fedorov _______________________________________________ slicer-devel mailing list slicer-devel at bwh.harvard.edu http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From curtislisle at knowledgevis.com Mon Apr 13 08:41:17 2009 From: curtislisle at knowledgevis.com (Curtis Lisle) Date: Mon, 13 Apr 2009 08:41:17 -0400 Subject: [slicer-devel] Saving DICOM header In-Reply-To: <81B5B66B035911478BFFDF078E0D810205176854@CINMLVEM14.e2k.ad.ge.com> References: <82f536810904081645n45133ad4wd2a6325bb0f26c38@mail.gmail.com> <81B5B66B035911478BFFDF078E0D810205176854@CINMLVEM14.e2k.ad.ge.com> Message-ID: I have also recently used the XNAT program's DicomBrowser to examine DICOM headers. It is available through the following URL: http://nrg.wustl.edu/projects/DICOM/DicomBrowser.jsp Curtis Lisle curtislisle at knowledgevis.com On Apr 13, 2009, at 8:32 AM, Miller, James V (GE, Research) wrote: > Hey Andriy, > > I usually use the tools from dcmtk, in particular dcmdump, to dump out > DICOM headers. > > > Jim Miller > Senior Scientist > GE Research > Visualization & Computer Vision > > GE imagination at work > > -----Original Message----- > From: slicer-devel-bounces at bwh.harvard.edu > [mailto:slicer-devel-bounces at bwh.harvard.edu] On Behalf Of Andriy > Fedorov > Sent: Wednesday, April 08, 2009 7:46 PM > To: Slicer Devel List > Subject: [slicer-devel] Saving DICOM header > > Hi, > > I am curious if there is a tool to extract and save (parts of) DICOM > header? I see Slicer is able to display DICOM header, but I would like > to store some of these (like acquisition parameters, patient > history) in > a text file. > > I assume I can find the DICOM parser part in Slicer, and make a > separate > tool, but maybe something like this is already available. > > Thanks > > Andriy Fedorov > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From nrannou at bwh.harvard.edu Mon Apr 13 12:33:25 2009 From: nrannou at bwh.harvard.edu (Nicolas RANNOU) Date: Mon, 13 Apr 2009 12:33:25 -0400 Subject: [slicer-devel] =?utf-8?q?_Fwd=3A_Re=3A_Error_=22Build=22_with_Ecl?= =?utf-8?q?ipse_after_=22make=22_with_Cmake2=2E6_-_VTC=5FWRAP=5FPYTHON?= Message-ID: <7f9cf8d835f1d2c70b989612d541241f@bwh.harvard.edu> -------- Original Message -------- Subject: Re: Error "Build" with Eclipse after "make" with Cmake2.6 - VTC_WRAP_PYTHON Date: Mon, 13 Apr 2009 12:27:55 -0400 From: Robert McKie To: Nicolas RANNOU Nicolas, This is an issue better handled by a note to "slicer-devel". We (meaning: help) don't get involved with software building, particularly slicer, and can't be of much assistance here. Thanks, Bob McKie Nicolas RANNOU wrote: > Hello, > > I'm currently trying to build Slicer 3 manually to then debugg it with > Eclipse but I have one issue. > I have a problem to Build VTK-5.4.0 > Everythin is fine with CMake and I can build the project using Eclipse if I > don't check VTK_WRAP_PYTHON in CMake(2.6 patch3) but I need it. > My version of Eclipse is the one which was originally on the computer. > > The error is the following: > > [ 96%] Converting Tcl test > cd /d/scratchbuild/nrannou/VTK-5.4.0-Debug/Filtering/Testing/Python && > /usr/bin/python2.5 > /d/scratchbuild/nrannou/VTK-5.4.0-Debug/Utilities/vtkTclTest2Py/vtkTclToPyConvertor.py > -l > /d/scratchbuild/nrannou/VTK-5.4.0-Debug/Filtering/Testing/Python/vtkTestsToConvert > -t > /d/scratchbuild/nrannou/VTK-5.4.0-Debug/Filtering/Testing/Python/conversion_complete > -A /d/scratchbuild/nrannou/VTK-5.4.0-Debug/Utilities/ > vtkTclTest2Py -A /d/scratchbuild/nrannou/VTK-5.4.0-Debug/Wrapping/Python -A > /d/scratchbuild/nrannou/VTK-5.4.0-Debug/bin/. > Converting > /projects/nacvisitors/nrannou/VTK-5.4.0/Filtering/Testing/Tcl/closedSplines.tcl > Conversion failed! > gmake[2]: Leaving directory `/d/scratchbuild/nrannou/VTK-5.4.0-Debug' > gmake[1]: Leaving directory `/d/scratchbuild/nrannou/VTK-5.4.0-Debug' > gmake[2]: *** [Filtering/Testing/Python/conversion_complete] Error 1 > gmake[1]: *** > [Filtering/Testing/Python/CMakeFiles/FilteringPythonTests.dir/all] Error 2 > gmake: *** [all] Error 2 > > > I'm building VTK in d/scratchbuild/nrannou/ as advised. > If I build it directly in my project folder the same error occurs. > > When I "make" it in CMake,the PYTHON_DEBUG_LIBRARY is NOTFOUND. > That may be the problem but I'm not very used to use Fedora and I can't fix > it. > Could you help me with this issue? > > Other people working in my lab did the same things but they could build the > project in Eclipse, even if python Library wasn't found. > They are not in projects/nacvisitors but project/na-mic/visitors > I don't know if it matters. > > Thank you, > Nicolas > > From harish.slicer at gmail.com Mon Apr 13 15:12:58 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 13 Apr 2009 12:12:58 -0700 Subject: [slicer-devel] slicer output Message-ID: Hi all, I try to put some print statements in my code and am able to print them in rxvt terminal. But is there any possibility to see these print statements in Slicer GUI. I dont see them in the log but is there any other source. Also, while debugging the slicer code, is print statements a good choice ? Though I am able to print the statements to terminal, I am not able to visualize the result. Hence I need to again open Slicer.exe and then execute the module which doesnt print the "print" statements. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090413/e0198df0/attachment.html From pieper at bwh.harvard.edu Mon Apr 13 15:38:57 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 13 Apr 2009 15:38:57 -0400 Subject: [slicer-devel] slicer output In-Reply-To: References: Message-ID: <49E394D1.1010403@bwh.harvard.edu> Hi Harish - If your code is in a vtk class, you can use the vtkErrorMacro to make your message show up in the Error Log window. For a command line module, the stdout shows up in the Error Log when the program completes. There is also a way for command line modules to print special codes to stdout that make messages show up in the status line of the application window; I don't remember the details off the top of my head, but you can look in some of the example Application/CLI code. Best, Steve Harish Doddi wrote: > > > Hi all, > > I try to put some print statements in my code and am able to print them > in rxvt terminal. But is there any possibility to see these print > statements in Slicer GUI. I dont see them in the log but is there any > other source. > > Also, while debugging the slicer code, is print statements a good choice > ? Though I am able to print the statements to terminal, I am not able to > visualize the result. Hence I need to again open Slicer.exe and then > execute the module which doesnt print the "print" statements. > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From harish.slicer at gmail.com Tue Apr 14 13:59:47 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Tue, 14 Apr 2009 10:59:47 -0700 Subject: [slicer-devel] model file coordinate system Message-ID: Hi all, I have a patella model in the format vtk. Is there any way to know the coordinate system followed in this model ie either LPS (Lateral Posterior Superior) or RAS(Right Anterior Superior). I have also attached the model for reference. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090414/299ce43a/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: 58_patella_right_center.vtk Type: application/octet-stream Size: 504436 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090414/299ce43a/attachment-0001.obj From pieper at bwh.harvard.edu Tue Apr 14 16:59:52 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 14 Apr 2009 16:59:52 -0400 Subject: [slicer-devel] update on bugs and release timing In-Reply-To: <49DBCF0E.6060606@bwh.harvard.edu> References: <49DBCF0E.6060606@bwh.harvard.edu> Message-ID: <49E4F948.4020503@bwh.harvard.edu> Update: after over 100 svn commits in the past week, we're now down to 8 bugs that are tagged for fix before the release - most of these are pretty small and/or just require confirmation before closing. Thanks, developers, for all the hard work! So we will be creating the 3.4 release branch tomorrow afternoon and start making release candidate builds. Please do not check in any changes unless they are on the bug list or reported to the list. Thanks, Steve Steve Pieper wrote: > Hi - > > After going through the bug list we've identified 19 bugs that are > tagged as "3.4 Targeted fix". The link should get you to them or you > can see the attached image. > > http://tinyurl.com/cu9bzp > > Our plan now is to create the 3.4 release branch on Tuesday April 14. > > ** Any fixes after that date will need to be manually checked into both > the trunk and the release branch. > > Please keep fixing bugs and testing the nightly builds. Add any new > issues to the bug tracker and flag them if you think they are critical > for the release. > > Thanks, > Steve > > ------------------------------------------------------------------------ > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Tue Apr 14 17:07:03 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 14 Apr 2009 17:07:03 -0400 Subject: [slicer-devel] [slicer-users] model file coordinate system In-Reply-To: References: Message-ID: <49E4FAF7.4010802@bwh.harvard.edu> Hi Harish - Model files for slicer have coordinates in RAS in millimeters. That's not really encoded in the .vtk file format, so it's by convention. -Steve Harish Doddi wrote: > > > Hi all, > > I have a patella model in the format vtk. Is there any way to know the > coordinate system followed in this model ie either LPS (Lateral > Posterior Superior) or RAS(Right Anterior Superior). I have also > attached the model for reference. > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From nrannou at bwh.harvard.edu Tue Apr 14 20:07:19 2009 From: nrannou at bwh.harvard.edu (Nicolas RANNOU) Date: Tue, 14 Apr 2009 20:07:19 -0400 Subject: [slicer-devel] Errors during getbuildtest.tcl Message-ID: Hello, I'm trying to install Slicer 3 with the "All in one script" doing : svn co http://svn.slicer.org/Slicer3/trunk Slicer3 ./Slicer3/Scripts/getbuildtest.tcl But at the end of the build, an error always occurs: ********************************************************************** ********************************************************************** The following tests FAILED: 11 - Slicer3GUITest1 (Failed) 12 - Slicer3GUITest2 (Failed) 13 - Slicer3FiducialsTest1 (Failed) 14 - Slicer3FiducialsTest2 (Failed) 15 - QdecModuleLibs (Failed) 16 - TractographyTestGlyphs (Failed) 20 - DiffusionEditorTest (Failed) 21 - DiffusionEditorGradients (Failed) 22 - DiffusionEditorGlyphsTracts (Failed) 23 - DiffusionEditorTensor (Failed) 40 - ModelIntoLabelVolumeTest (Failed) 65 - ExecutionModelTourTest (Failed) 90 - Slicer3CLTest5 (Failed) 91 - Slicer3CLTest6 (Failed) 92 - Slicer3CLTest7 (Failed) 93 - Slicer3ScrollTest (Failed) Performing coverage Cannot find any coverage files. Ignoring Coverage request. Submit files (using http) Using HTTP submit method Drop site: http://www.cdash.org/CDash/submit.php?project=Slicer3 Uploaded: /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Build.xml Uploaded: /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Configure.xml Uploaded: /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Test.xml Submission successful Errors while running CTest make[3]: *** [CMakeFiles/Experimental] Error 8 make[2]: *** [CMakeFiles/Experimental.dir/all] Error 2 make[1]: *** [CMakeFiles/Experimental.dir/rule] Error 2 make: *** [Experimental] Error 2 child process exited abnormally Results: build of "Experimental" failed ************************************************************************ ************************************************************************ I can launch Slicer 3 and do the tutorials but I must solve this problem. Do you have any idea about how I could do? Please find enclosed a more precise description of what I've got my termninal if it can help. Thanks, Nicolas -------------- next part -------------- -- Performing Curl Test HAVE_LONG_LONG_CONSTANT -- Performing Curl Test HAVE_LONG_LONG_CONSTANT - Success -- Check size of curl_off_t -- Check size of curl_off_t - failed -- Configuring done -- Generating done -- Build files have been written to: /d/scratchbuild/nrannou/SPL/Slicer3-lib/cmcurl-build running: make -j 16 Scanning dependencies of target slicerlibcurl [ 2%] [ 6%] [ 6%] [ 8%] [ 10%] [ 14%] [ 14%] [ 18%] [ 18%] [ 25%] [ 27%] [ 27%] [ 29%] [ 29%] Building C object CMakeFiles/slicerlibcurl.dir/base64.o [ 33%] [ 33%] Building C object CMakeFiles/slicerlibcurl.dir/content_encoding.o Building C object CMakeFiles/slicerlibcurl.dir/connect.o Building C object CMakeFiles/slicerlibcurl.dir/cookie.o Building C object CMakeFiles/slicerlibcurl.dir/escape.o Building C object CMakeFiles/slicerlibcurl.dir/dict.o Building C object CMakeFiles/slicerlibcurl.dir/easy.o Building C object CMakeFiles/slicerlibcurl.dir/file.o Building C object CMakeFiles/slicerlibcurl.dir/formdata.o Building C object CMakeFiles/slicerlibcurl.dir/hash.o Building C object CMakeFiles/slicerlibcurl.dir/ftp.o Building C object CMakeFiles/slicerlibcurl.dir/getenv.o Building C object CMakeFiles/slicerlibcurl.dir/getdate.o Building C object CMakeFiles/slicerlibcurl.dir/getinfo.o Building C object CMakeFiles/slicerlibcurl.dir/hostares.o Building C object CMakeFiles/slicerlibcurl.dir/hostasyn.o [ 35%] [ 37%] [ 39%] [ 43%] [ 43%] Building C object CMakeFiles/slicerlibcurl.dir/hostip.o [ 45%] [ 50%] [ 50%] [ 52%] [ 56%] [ 56%] Building C object CMakeFiles/slicerlibcurl.dir/hostip6.o [ 58%] Building C object CMakeFiles/slicerlibcurl.dir/hostip4.o [ 60%] Building C object CMakeFiles/slicerlibcurl.dir/hostsyn.o Building C object CMakeFiles/slicerlibcurl.dir/hostthre.o [ 62%] Building C object CMakeFiles/slicerlibcurl.dir/http.o [ 64%] Building C object CMakeFiles/slicerlibcurl.dir/http_digest.o Building C object CMakeFiles/slicerlibcurl.dir/http_chunks.o Building C object CMakeFiles/slicerlibcurl.dir/http_negotiate.o Building C object CMakeFiles/slicerlibcurl.dir/http_ntlm.o Building C object CMakeFiles/slicerlibcurl.dir/inet_ntop.o Building C object CMakeFiles/slicerlibcurl.dir/if2ip.o [ 66%] Building C object CMakeFiles/slicerlibcurl.dir/inet_pton.o Building C object CMakeFiles/slicerlibcurl.dir/krb4.o Building C object CMakeFiles/slicerlibcurl.dir/llist.o Building C object CMakeFiles/slicerlibcurl.dir/md5.o [ 68%] [ 70%] [ 75%] [ 75%] [ 79%] [ 79%] Building C object CMakeFiles/slicerlibcurl.dir/memdebug.o [ 81%] Building C object CMakeFiles/slicerlibcurl.dir/mprintf.o [ 83%] [ 85%] [ 89%] [ 89%] Building C object CMakeFiles/slicerlibcurl.dir/multi.o [ 91%] Building C object CMakeFiles/slicerlibcurl.dir/sendf.o Building C object CMakeFiles/slicerlibcurl.dir/netrc.o Building C object CMakeFiles/slicerlibcurl.dir/progress.o [ 95%] [ 97%] [ 97%] Building C object CMakeFiles/slicerlibcurl.dir/share.o Building C object CMakeFiles/slicerlibcurl.dir/speedcheck.o [100%] Building C object CMakeFiles/slicerlibcurl.dir/ssluse.o Building C object CMakeFiles/slicerlibcurl.dir/strerror.o Building C object CMakeFiles/slicerlibcurl.dir/strequal.o Building C object CMakeFiles/slicerlibcurl.dir/telnet.o Building C object CMakeFiles/slicerlibcurl.dir/timeval.o Building C object CMakeFiles/slicerlibcurl.dir/transfer.o Building C object CMakeFiles/slicerlibcurl.dir/url.o Building C object CMakeFiles/slicerlibcurl.dir/version.o Linking C static library bin/libslicerlibcurl.a [100%] Built target slicerlibcurl All packages compiled. running: /d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/CMake-build/bin/cmake {-GUnix Makefiles} {-DMAKECOMMAND:STRING=make -j 16} -DCMAKE_CXX_COMPILER:STRING=/usr/bin/g++ -DCMAKE_CXX_COMPILER_FULLPATH:FILEPATH=/usr/bin/g++ -DITK_DIR:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/Insight-build -DKWWidgets_DIR:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/KWWidgets-build -DTeem_DIR:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/teem-build -DOpenIGTLink_DIR:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/OpenIGTLink-build -DBatchMake_DIR:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/BatchMake-build -DSlicer3_USE_BatchMake=ON -DINCR_TCL_LIBRARY:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/tcl-build/lib/lib/libitcl3.2.so -DINCR_TK_LIBRARY:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/tcl-build/lib/lib/libitk3.2.so -DSlicer3_USE_PYTHON=ON -DSlicer3_USE_NUMPY=ON -DSlicer3_USE_OPENIGTLINK=ON -DPYTHON_INCLUDE_PATH:PATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/python-build/include/python2.5 -DPYTHON_LIBRARY:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/python-build/lib/libpython2.5.so -DSandBox_DIR:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/NAMICSandBox -DCMAKE_BUILD_TYPE=Debug -DSlicer3_VERSION_PATCH:STRING=2009-04-14 -DCPACK_GENERATOR:STRING=TGZ -DCPACK_PACKAGE_FILE_NAME:STRING=Slicer3-3.3-alpha-2009-04-14-linux-x86_64 -DCPACK_RPM_SPEC_MORE_DEFINE= -DSLICERLIBCURL_DIR:FILEPATH=/d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/cmcurl-build -DCMAKE_VERBOSE_MAKEFILE:BOOL=OFF /d/scratchbuild/nrannou/SPL/Slicer3 -- The C compiler identification is GNU -- The CXX compiler identification is GNU -- Check for working C compiler: /usr/bin/gcc -- Check for working C compiler: /usr/bin/gcc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Check for working CXX compiler: /usr/bin/g++ -- Check for working CXX compiler: /usr/bin/g++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Check if the system is big endian -- Searching 16 bit integer -- Looking for sys/types.h -- Looking for sys/types.h - found -- Looking for stdint.h -- Looking for stdint.h - found -- Looking for stddef.h -- Looking for stddef.h - found -- Check size of unsigned short -- Check size of unsigned short - done -- Using unsigned short -- Check if the system is big endian - little endian -- Found PythonLibs: /d/scratchbuild/nrannou/SPL/Slicer3/../Slicer3-lib/python-build/lib/libpython2.5.so -- Configuring done -- Generating done -- Build files have been written to: /d/scratchbuild/nrannou/SPL/Slicer3-build running: make -j 16 Experimental Scanning dependencies of target Experimental Site: b2_tm128.bwh.harvard.edu Build name: Linux-g++ Create new tag: 20090414-2134 - Experimental Start processing tests Configure project Each . represents 1024 bytes of output . Size of output: 0K Build project Each symbol represents 1024 bytes of output. '!' represents an error and '*' a warning. .................................................. Size: 50K ...........***........**.*********................ Size: 100K ............**.................................... Size: 150K ...........*...................................... Size: 200K ..........*....................................... Size: 250K .............................*.................... Size: 300K ............. Size of output: 312K 0 Compiler errors 29 Compiler warnings Test project /d/scratchbuild/nrannou/SPL/Slicer3-build 1/ 96 Testing Parser1Test1 Passed 2/ 96 Testing Parser1Test2 Passed 3/ 96 Testing ModuleFactoryTest Passed 4/ 96 Testing Slicer3CLPExample1 Passed 5/ 96 Testing Slicer3CLPExample2 Passed 6/ 96 Testing Slicer3CLPExample3 Passed 7/ 96 Testing LMParser1Test1 Passed 8/ 96 Testing LMParser1Test2 Passed 9/ 96 Testing VTKITKVectorReader Passed 10/ 96 Testing VTKITKBSplineTransform Passed 11/ 96 Testing Slicer3GUITest1 ***Failed 12/ 96 Testing Slicer3GUITest2 ***Failed 13/ 96 Testing Slicer3FiducialsTest1 ***Failed 14/ 96 Testing Slicer3FiducialsTest2 ***Failed 15/ 96 Testing QdecModuleLibs ***Failed 16/ 96 Testing TractographyTestGlyphs ***Failed 17/ 96 Testing ChangeTrackerCL_Help Passed 18/ 96 Testing ChangeTrackerCL_TestGlobalReg Passed 19/ 96 Testing ChangeTrackerCL_TestFull_Inten Passed 20/ 96 Testing DiffusionEditorTest ***Failed 21/ 96 Testing DiffusionEditorGradients ***Failed 22/ 96 Testing DiffusionEditorGlyphsTracts ***Failed 23/ 96 Testing DiffusionEditorTensor ***Failed 24/ 96 Testing vtkEMSegmentMRMLManagerTest Passed 25/ 96 Testing vtkEMSegmentPreprocessingManag Passed 26/ 96 Testing vtkEMSegmentBlackBoxSegmentati Passed 27/ 96 Testing vtkEMSegmentBuildAndRunNewSegm Passed 28/ 96 Testing EMSegCL_Help Passed 29/ 96 Testing EMSegCL_Version Passed 30/ 96 Testing EMSegCL_RunDefaultNodes Passed 31/ 96 Testing EMSegCL_RunNormalize Passed 32/ 96 Testing EMSegCL_RunSetEverything Passed 33/ 96 Testing EMSegCL_DisableMultithreading Passed 34/ 96 Testing EMSegCL_EFBogusParameterNode Passed 35/ 96 Testing EMSegCL_EFBogusMRMLScene Passed 36/ 96 Testing EMSegCL_EFBogusTargetImages Passed 37/ 96 Testing EMSegCL_EFTooManyImages Passed 38/ 96 Testing EMSegCL_EFTooFewImages Passed 39/ 96 Testing EMSegCL_EFBogusResultStandardI Passed 40/ 96 Testing ModelIntoLabelVolumeTest ***Failed 41/ 96 Testing OrientImageTest1 Passed 42/ 96 Testing OrientImageTest2 Passed 43/ 96 Testing OrientImageTest3 Passed 44/ 96 Testing AddTest Passed 45/ 96 Testing ThresholdTest Passed 46/ 96 Testing CheckerBoardTest Passed 47/ 96 Testing MedianImageFilterTest Passed 48/ 96 Testing GaussianBlurImageFilterTest Passed 49/ 96 Testing LinearRegistrationTest00 Passed 50/ 96 Testing LinearRegistrationTest01 Passed 51/ 96 Testing LinearRegistrationTest02 Passed 52/ 96 Testing LinearRegistrationTest03 Passed 53/ 96 Testing LinearRegistrationTest04 Passed 54/ 96 Testing BSplineRegistrationTest00 Passed 55/ 96 Testing TestGridTransformRegistrationT Passed 56/ 96 Testing AffineRegistrationTest Passed 57/ 96 Testing ConfidenceConnectedTest Passed 58/ 96 Testing CurvatureAnisotropicDiffusionT Passed 59/ 96 Testing DicomToNrrdConverterTest Passed 60/ 96 Testing DiffusionTensorEstimationTest Passed 61/ 96 Testing DiffusionTensorMathematicsTest Passed 62/ 96 Testing DiffusionTensorTestTest Passed 63/ 96 Testing DiffusionWeightedTestTest Passed 64/ 96 Testing DWIDicomLoadTest Passed 65/ 96 Testing ExecutionModelTourTest ***Failed 66/ 96 Testing FreesurferSurfaceSectionExtrac Passed 67/ 96 Testing GradientAnisotropicDiffusionTe Passed 68/ 96 Testing GrayscaleFillHoleImageFilterTe Passed 69/ 96 Testing GrayscaleGrindPeakImageFilterT Passed 70/ 96 Testing GrayscaleModelMakerTest Passed 71/ 96 Testing HistogramMatchingTest Passed 72/ 96 Testing ImageLabelCombineTest Passed 73/ 96 Testing ImageReadDicomWriteTest Passed 74/ 96 Testing LabelMapSmoothingTest Passed 75/ 96 Testing ModelMakerTest Passed 76/ 96 Testing MultipleModelsExampleTest Passed 77/ 96 Testing OtsuThresholdImageFilterTest Passed 78/ 96 Testing OtsuThresholdSegmentationTest Passed 79/ 96 Testing ResampleVolume2Test Passed 80/ 96 Testing ResampleVolumeTest Passed 81/ 96 Testing RigidRegistrationTest01 Passed 82/ 96 Testing RigidRegistrationTest02 Passed 83/ 96 Testing SubtractTest Passed 84/ 96 Testing VotingBinaryHoleFillingImageFi Passed 85/ 96 Testing ZeroCrossingBasedEdgeDetection Passed 86/ 96 Testing Slicer3CLTest1 Passed 87/ 96 Testing Slicer3CLTest2 Passed 88/ 96 Testing Slicer3CLTest3 Passed 89/ 96 Testing Slicer3CLTest4 Passed 90/ 96 Testing Slicer3CLTest5 ***Failed 91/ 96 Testing Slicer3CLTest6 ***Failed 92/ 96 Testing Slicer3CLTest7 ***Failed 93/ 96 Testing Slicer3ScrollTest ***Failed 94/ 96 Testing Slicer3MRMLUndo Passed 95/ 96 Testing Slicer3MRMLVolume Passed 96/ 96 Testing EventBrokerTest Passed 83% tests passed, 16 tests failed out of 96 The following tests FAILED: 11 - Slicer3GUITest1 (Failed) 12 - Slicer3GUITest2 (Failed) 13 - Slicer3FiducialsTest1 (Failed) 14 - Slicer3FiducialsTest2 (Failed) 15 - QdecModuleLibs (Failed) 16 - TractographyTestGlyphs (Failed) 20 - DiffusionEditorTest (Failed) 21 - DiffusionEditorGradients (Failed) 22 - DiffusionEditorGlyphsTracts (Failed) 23 - DiffusionEditorTensor (Failed) 40 - ModelIntoLabelVolumeTest (Failed) 65 - ExecutionModelTourTest (Failed) 90 - Slicer3CLTest5 (Failed) 91 - Slicer3CLTest6 (Failed) 92 - Slicer3CLTest7 (Failed) 93 - Slicer3ScrollTest (Failed) Performing coverage Cannot find any coverage files. Ignoring Coverage request. Submit files (using http) Using HTTP submit method Drop site: http://www.cdash.org/CDash/submit.php?project=Slicer3 Uploaded: /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Build.xml Uploaded: /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Configure.xml Uploaded: /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Test.xml Submission successful Errors while running CTest make[3]: *** [CMakeFiles/Experimental] Error 8 make[2]: *** [CMakeFiles/Experimental.dir/all] Error 2 make[1]: *** [CMakeFiles/Experimental.dir/rule] Error 2 make: *** [Experimental] Error 2 child process exited abnormally Results: build of "Experimental" failed From harish.slicer at gmail.com Tue Apr 14 21:32:45 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Tue, 14 Apr 2009 18:32:45 -0700 Subject: [slicer-devel] colors in linux Message-ID: Hi all, I am using slicer in Linux operating system. In the models module, I try to give different colors to various models. However though I could select the colors, there is no "OK" button present as a result I am not able to assign a color to my model. Is there any alternative. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090414/c055e90c/attachment-0001.html From pieper at bwh.harvard.edu Wed Apr 15 08:20:01 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 15 Apr 2009 08:20:01 -0400 Subject: [slicer-devel] Errors during getbuildtest.tcl In-Reply-To: References: Message-ID: <49E5D0F1.4000804@bwh.harvard.edu> Hi Nicolas - Unfortunately some of the tests failing is 'normal' and is typically due to memory leaks on exit. This shouldn't get in the way of using the build. -Steve Nicolas RANNOU wrote: > Hello, > > I'm trying to install Slicer 3 with the "All in one script" doing : > > > svn co http://svn.slicer.org/Slicer3/trunk Slicer3 > > ./Slicer3/Scripts/getbuildtest.tcl > > > But at the end of the build, an error always occurs: > > > ********************************************************************** > ********************************************************************** > > The following tests FAILED: > 11 - Slicer3GUITest1 (Failed) > 12 - Slicer3GUITest2 (Failed) > 13 - Slicer3FiducialsTest1 (Failed) > 14 - Slicer3FiducialsTest2 (Failed) > 15 - QdecModuleLibs (Failed) > 16 - TractographyTestGlyphs (Failed) > 20 - DiffusionEditorTest (Failed) > 21 - DiffusionEditorGradients (Failed) > 22 - DiffusionEditorGlyphsTracts (Failed) > 23 - DiffusionEditorTensor (Failed) > 40 - ModelIntoLabelVolumeTest (Failed) > 65 - ExecutionModelTourTest (Failed) > 90 - Slicer3CLTest5 (Failed) > 91 - Slicer3CLTest6 (Failed) > 92 - Slicer3CLTest7 (Failed) > 93 - Slicer3ScrollTest (Failed) > Performing coverage > Cannot find any coverage files. Ignoring Coverage request. > Submit files (using http) > Using HTTP submit method > Drop site: http://www.cdash.org/CDash/submit.php?project=Slicer3 > Uploaded: > /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Build.xml > Uploaded: > /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Configure.xml > Uploaded: > /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/Test.xml > Submission successful > Errors while running CTest > make[3]: *** [CMakeFiles/Experimental] Error 8 > make[2]: *** [CMakeFiles/Experimental.dir/all] Error 2 > make[1]: *** [CMakeFiles/Experimental.dir/rule] Error 2 > make: *** [Experimental] Error 2 > > child process exited abnormally > > Results: > build of "Experimental" failed > > ************************************************************************ > ************************************************************************ > > > I can launch Slicer 3 and do the tutorials but I must solve this problem. > Do you have any idea about how I could do? > Please find enclosed a more precise description of what I've got my > termninal if it can help. > > > Thanks, > Nicolas > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From simon at math.uh.edu Wed Apr 15 08:53:29 2009 From: simon at math.uh.edu (Simon Alexander) Date: Wed, 15 Apr 2009 07:53:29 -0500 Subject: [slicer-devel] Errors during getbuildtest.tcl In-Reply-To: References: Message-ID: <8944E10E-88D2-4E3E-88F6-939CD3AEC8EE@math.uh.edu> Hi Nicolas, Are you running on OS X? In that case I've noticed a similar set of errors if there is an issue with finding the X11 display. In that case, what happens if you run the getbuildtest.tcl (in a clean X11 if you've been changing networks) like DISPLAY=:0 ./Slicer3/Scripts/getbuildtest.tcl instead? cheers, simon On Apr 14, 2009, at 7:07 PM, Nicolas RANNOU wrote: > Hello, > > I'm trying to install Slicer 3 with the "All in one script" doing : > > > svn co http://svn.slicer.org/Slicer3/trunk Slicer3 > > ./Slicer3/Scripts/getbuildtest.tcl > > > But at the end of the build, an error always occurs: > > > ********************************************************************** > ********************************************************************** > > The following tests FAILED: > 11 - Slicer3GUITest1 (Failed) > 12 - Slicer3GUITest2 (Failed) > 13 - Slicer3FiducialsTest1 (Failed) > 14 - Slicer3FiducialsTest2 (Failed) > 15 - QdecModuleLibs (Failed) > 16 - TractographyTestGlyphs (Failed) > 20 - DiffusionEditorTest (Failed) > 21 - DiffusionEditorGradients (Failed) > 22 - DiffusionEditorGlyphsTracts (Failed) > 23 - DiffusionEditorTensor (Failed) > 40 - ModelIntoLabelVolumeTest (Failed) > 65 - ExecutionModelTourTest (Failed) > 90 - Slicer3CLTest5 (Failed) > 91 - Slicer3CLTest6 (Failed) > 92 - Slicer3CLTest7 (Failed) > 93 - Slicer3ScrollTest (Failed) > Performing coverage > Cannot find any coverage files. Ignoring Coverage request. > Submit files (using http) > Using HTTP submit method > Drop site: http://www.cdash.org/CDash/submit.php?project=Slicer3 > Uploaded: > /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/ > Build.xml > Uploaded: > /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/ > Configure.xml > Uploaded: > /d/scratchbuild/nrannou/SPL/Slicer3-build/Testing/20090414-2134/ > Test.xml > Submission successful > Errors while running CTest > make[3]: *** [CMakeFiles/Experimental] Error 8 > make[2]: *** [CMakeFiles/Experimental.dir/all] Error 2 > make[1]: *** [CMakeFiles/Experimental.dir/rule] Error 2 > make: *** [Experimental] Error 2 > > child process exited abnormally > > Results: > build of "Experimental" failed > > ************************************************************************ > ************************************************************************ > > > I can launch Slicer 3 and do the tutorials but I must solve this > problem. > Do you have any idea about how I could do? > Please find enclosed a more precise description of what I've got my > termninal if it can help. > > > Thanks, > Nicolas_______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From nicole at bwh.harvard.edu Wed Apr 15 10:10:03 2009 From: nicole at bwh.harvard.edu (Nicole Aucoin) Date: Wed, 15 Apr 2009 10:10:03 -0400 (EDT) Subject: [slicer-devel] colors in linux In-Reply-To: References: Message-ID: Hello Harish, Could you please send me a screen shot of what the Color Picker pop up looks like when you click on the Set Color button? I'm always using Linux and haven't seen the Ok or Cancel buttons disappear - there're there as of last night's build. The procedural way to do it is to turn on the MRML node ids in the data module, make note of your model's id, and then enter the following command in the tkcon terminal (to turn the first model red): [[$::slicer3::MRMLScene GetNodeByID vtkMRMLModelNode4] GetDisplayNode] SetColor 1.0 0.0 0.0 Nicole On Tue, 14 Apr 2009, Harish Doddi wrote: > Hi all, > > I am using slicer in Linux operating system. In the models module, I try to > give different colors to various models. However though I could select the > colors, there is no "OK" button present as a result I am not able to assign > a color to my model. Is there any alternative. > > Thanks, > Harish > -- Nicole Aucoin nicole at bwh.harvard.edu Surgical Planning Laboratory, Brigham and Women's Hospital Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 http://www.slicer.org From harish.slicer at gmail.com Wed Apr 15 12:06:13 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 15 Apr 2009 09:06:13 -0700 Subject: [slicer-devel] colors in linux In-Reply-To: References: Message-ID: Hi Nicole, Attached is a screen shot of the color picker in which there is no "OK" button. Thanks, Harish On 4/15/09, Nicole Aucoin wrote: > > Hello Harish, > > Could you please send me a screen shot of what the Color Picker pop up > looks like when you click on the Set Color button? I'm always using Linux > and haven't seen the Ok or Cancel buttons disappear - there're there as of > last night's build. > > The procedural way to do it is to turn on the MRML node ids in the data > module, make note of your model's id, and then enter the following command > in the tkcon terminal (to turn the first model red): > [[$::slicer3::MRMLScene GetNodeByID vtkMRMLModelNode4] GetDisplayNode] > SetColor 1.0 0.0 0.0 > > Nicole > > On Tue, 14 Apr 2009, Harish Doddi wrote: > > Hi all, >> >> I am using slicer in Linux operating system. In the models module, I try >> to >> give different colors to various models. However though I could select the >> colors, there is no "OK" button present as a result I am not able to >> assign >> a color to my model. Is there any alternative. >> >> Thanks, >> Harish >> >> > -- > Nicole Aucoin nicole at bwh.harvard.edu > Surgical Planning Laboratory, Brigham and Women's Hospital > Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 > http://www.slicer.org > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090415/268ded04/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-3D Slicer Version 3.3 Alpha: Color Picker.tiff Type: image/tiff Size: 36355 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090415/268ded04/attachment-0001.tiff From harish.slicer at gmail.com Wed Apr 15 12:08:35 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 15 Apr 2009 09:08:35 -0700 Subject: [slicer-devel] colors in linux In-Reply-To: References: Message-ID: Hi Nicole, Attached is a screen shot of the color picker in which there is no "OK" button. Thanks, Harish On 4/15/09, Nicole Aucoin wrote: > > Hello Harish, > > Could you please send me a screen shot of what the Color Picker pop up > looks like when you click on the Set Color button? I'm always using Linux > and haven't seen the Ok or Cancel buttons disappear - there're there as of > last night's build. > > The procedural way to do it is to turn on the MRML node ids in the data > module, make note of your model's id, and then enter the following command > in the tkcon terminal (to turn the first model red): > [[$::slicer3::MRMLScene GetNodeByID vtkMRMLModelNode4] GetDisplayNode] > SetColor 1.0 0.0 0.0 > > Nicole > > On Tue, 14 Apr 2009, Harish Doddi wrote: > > Hi all, >> >> I am using slicer in Linux operating system. In the models module, I try >> to >> give different colors to various models. However though I could select the >> colors, there is no "OK" button present as a result I am not able to >> assign >> a color to my model. Is there any alternative. >> >> Thanks, >> Harish >> >> > -- > Nicole Aucoin nicole at bwh.harvard.edu > Surgical Planning Laboratory, Brigham and Women's Hospital > Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 > http://www.slicer.org > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090415/1ea40e7a/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-3D Slicer Version 3.3 Alpha: Color Picker.tiff Type: image/tiff Size: 36355 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090415/1ea40e7a/attachment-0001.tiff From pieper at bwh.harvard.edu Wed Apr 15 12:21:52 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 15 Apr 2009 12:21:52 -0400 Subject: [slicer-devel] prostatenav crashing on start up In-Reply-To: References: Message-ID: <49E609A0.5010802@bwh.harvard.edu> Hi Nicole - I disabled ProstateNav in Modules/CMakeLists.txt yesterday due to other crashes - so it's not being rebuilt when you run make. You need to delete the libs from the build directory or they will be discovered and loaded incorrectly. Haiying is going to work on the original issue tomorrow and hopefully we can re-enable the module. -Steve Nicole Aucoin wrote: > I can't quite figure out where the problem is coming from due to the > superclass triggering the crash, any ideas? > > 0x00002aaaac2e8a24 in vtkDataArrayTemplate::GetValue > (this=0x2aaaae0c3f1c, id=0) > at > /projects/birn/nicole/Linux/Slicer3-lib/VTK/Common/vtkDataArrayTemplate.h:110 > > 110 T GetValue(vtkIdType id) { return this->Array[id]; } > (gdb) bt > #0 0x00002aaaac2e8a24 in vtkDataArrayTemplate::GetValue > (this=0x2aaaae0c3f1c, id=0) > at > /projects/birn/nicole/Linux/Slicer3-lib/VTK/Common/vtkDataArrayTemplate.h:110 > > #1 0x00002aaaac2e8a45 in vtkIntArray::GetValue (this=0x2aaaae0c3f1c, > id=0) at > /projects/birn/nicole/Linux/Slicer3-lib/VTK/Common/vtkIntArray.h:71 > #2 0x00002aaaaca77c19 in vtkObserverManager::AddObjectEvents > (this=0x20c7920, nodePtr=0x9b9520, events=0x2aaaae0c3f1c) > at > /projects/birn/nicole/Linux/Slicer3/Libs/MRML/vtkObserverManager.cxx:194 > #3 0x00002aaaaca78062 in vtkObserverManager::SetAndObserveObjectEvents > (this=0x20c7920, nodePtr=0x20c7740, node=0x9b9520, events=0x2aaaae0c3f1c) > at > /projects/birn/nicole/Linux/Slicer3/Libs/MRML/vtkObserverManager.cxx:139 > #4 0x000000000043eab4 in > vtkSlicerComponentGUI::SetAndObserveMRMLSceneEvents (this=0x20c7700, > mrml=0x9b9520, events=0x2aaaae0c3f1c) > at > /projects/birn/nicole/Linux/Slicer3/Base/GUI/vtkSlicerComponentGUI.h:120 > #5 0x00000000004266b5 in Slicer3_main (argc=1, argv=0x7fff946512b8) at > /projects/birn/nicole/Linux/Slicer3/Applications/GUI/Slicer3.cxx:1191 > #6 0x000000000042b8f7 in main (argc=1, argv=0x7fff946512b8) at > /projects/birn/nicole/Linux/Slicer3/Applications/GUI/Slicer3.cxx:2276 > > > From sebastien.barre at kitware.com Wed Apr 15 12:59:44 2009 From: sebastien.barre at kitware.com (Sebastien BARRE) Date: Wed, 15 Apr 2009 12:59:44 -0400 Subject: [slicer-devel] colors in linux In-Reply-To: References: Message-ID: <20090415170000.487A753FD@public.kitware.com> Hi Harish At 4/15/2009 12:08 PM, Harish Doddi wrote: >Hi Nicole, > > >Attached is a screen shot of the color picker in which there is no >"OK" button. The default font here seems huge, way bigger than the standard one (see window title). Did you change your font size? This dialog can't adjust itself properly at the moment (god knows I tried). But then again, you can't expect all dialogs to react properly to very large font size variation, they *will* break ultimately. From haehn at bwh.harvard.edu Wed Apr 15 13:02:44 2009 From: haehn at bwh.harvard.edu (Daniel Haehn) Date: Wed, 15 Apr 2009 13:02:44 -0400 Subject: [slicer-devel] vtkMRMLModelDisplayNode SetSliceIntersectionVisibility has no effect Message-ID: <49E61334.3000309@bwh.harvard.edu> Hi, I am creating a vtkMRMLModelNode and a vtkMRMLModelDisplayNode during execution of my python scripted module and SetSliceIntersectionVisibility(1) with no effect. When I save the generated model and load it again using the standard Slicer options, the SliceIntersectionVisibility setting of the Model module works great. So I assume I am doing something wrong during the creation of the nodes. Here is the python code: if self._mainGUIClass._outInitModel == None: # no node so far self._mainGUIClass._outInitModel = slicer.vtkMRMLModelNode() self._mainGUIClass._outInitModel.SetName("VMTK Level-Set Initialization Output Model") self._mainGUIClass._outInitModel.SetAndObservePolyData(slicer.vtkPolyData()) self._mainGUIClass._outInitModel.SetScene(scene) scene.AddNode(self._mainGUIClass._outInitModel) self._mainGUIClass._outInitModel.SetAndObservePolyData(polyData) self._mainGUIClass._outInitModel.SetModifiedSinceRead(1) if self._mainGUIClass._outInitModelDisplay!=None: scene.RemoveNode(self._mainGUIClass._outInitModelDisplay) self._mainGUIClass._outInitModelDisplay = slicer.vtkMRMLModelDisplayNode() self._mainGUIClass._outInitModelDisplay.SetPolyData(self._mainGUIClass._outInitModel.GetPolyData()) self._mainGUIClass._outInitModelDisplay.SetColor(0.8, 0.0, 0.0) self._mainGUIClass._outInitModelDisplay.SetSliceIntersectionVisibility(1) self._mainGUIClass._outInitModelDisplay.SetVisibility(1) self._mainGUIClass._outInitModelDisplay.SetOpacity(0.5) scene.AddNode(self._mainGUIClass._outInitModelDisplay) self._mainGUIClass._outInitModel.SetAndObserveDisplayNodeID(self._mainGUIClass._outInitModelDisplay.GetID()) I always re-add the vtkMRMLModelDisplayNode because else it would not be updated when it is changed. Maybe that is the issue? Is there another solution? Thank you very much! Daniel From harish.slicer at gmail.com Wed Apr 15 13:06:25 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 15 Apr 2009 10:06:25 -0700 Subject: [slicer-devel] colors in linux In-Reply-To: <20090415170000.487A753FD@public.kitware.com> References: <20090415170000.487A753FD@public.kitware.com> Message-ID: Hi Sebastien, I see the font settings dialog box in the view -> application settings -> font settings. The options I see are small, medium and large font sizes and already small is selected. I am still not able to see the "OK" window. Is there something I am missing? Thanks, Harish On 4/15/09, Sebastien BARRE wrote: > > Hi Harish > > At 4/15/2009 12:08 PM, Harish Doddi wrote: > >> Hi Nicole, >> >> >> Attached is a screen shot of the color picker in which there is no "OK" >> button. >> > > The default font here seems huge, way bigger than the standard one (see > window title). > Did you change your font size? > > This dialog can't adjust itself properly at the moment (god knows I tried). > But then again, you can't expect all dialogs to react properly to very large > font size variation, they *will* break ultimately. > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090415/96a1949c/attachment.html From sebastien.barre at kitware.com Wed Apr 15 13:09:29 2009 From: sebastien.barre at kitware.com (Sebastien BARRE) Date: Wed, 15 Apr 2009 13:09:29 -0400 Subject: [slicer-devel] colors in linux In-Reply-To: References: <20090415170000.487A753FD@public.kitware.com> Message-ID: <20090415170944.EB9644A8@public.kitware.com> At 4/15/2009 01:06 PM, Harish Doddi wrote: >Hi Sebastien, > >I see the font settings dialog box in the view -> application >settings -> font settings. The options I see are small, medium and >large font sizes and already small is selected. I am still not able >to see the "OK" window. Is there something I am missing? Were the fonts always that big in Slicer3 for you? Is this something that started recently? I assume the rest of the GUI must be pretty much unusable because of this size. Are you using it on Linux, or from a virtual machine like VMWare? From harish.slicer at gmail.com Wed Apr 15 13:24:49 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 15 Apr 2009 10:24:49 -0700 Subject: [slicer-devel] colors in linux In-Reply-To: <20090415170944.EB9644A8@public.kitware.com> References: <20090415170000.487A753FD@public.kitware.com> <20090415170944.EB9644A8@public.kitware.com> Message-ID: The font is the same every time. The font seems ok for the rest of the GUI. I am using it on Linux and am using Slicer 3.3. Is there any feature related to this recently incorporated in slicer? Thanks, Harish On Wed, Apr 15, 2009 at 10:09 AM, Sebastien BARRE < sebastien.barre at kitware.com> wrote: > At 4/15/2009 01:06 PM, Harish Doddi wrote: > >> Hi Sebastien, >> >> I see the font settings dialog box in the view -> application settings -> >> font settings. The options I see are small, medium and large font sizes and >> already small is selected. I am still not able to see the "OK" window. Is >> there something I am missing? >> > > Were the fonts always that big in Slicer3 for you? Is this something that > started recently? I assume the rest of the GUI must be pretty much unusable > because of this size. Are you using it on Linux, or from a virtual machine > like VMWare? > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090415/64703f54/attachment.html From pieper at bwh.harvard.edu Wed Apr 15 13:39:05 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 15 Apr 2009 13:39:05 -0400 Subject: [slicer-devel] colors in linux In-Reply-To: References: <20090415170000.487A753FD@public.kitware.com> <20090415170944.EB9644A8@public.kitware.com> Message-ID: <49E61BB9.7070904@bwh.harvard.edu> Ron also saw this on one of his linux machines, but it doesn't seem to occur on all machines. As far as Wendy or I know there was nothing in slicer that changed to impact this. Perhaps there's a linux font package that changed? Harish, what distribution do you use? I tested on ubuntu 8.10 and didn't have a problem. -Steve Harish Doddi wrote: > The font is the same every time. The font seems ok for the rest of the > GUI. I am using it on Linux and am using Slicer 3.3. Is there any > feature related to this recently incorporated in slicer? > > Thanks, > Harish > > On Wed, Apr 15, 2009 at 10:09 AM, Sebastien BARRE > > wrote: > > At 4/15/2009 01:06 PM, Harish Doddi wrote: > > Hi Sebastien, > > I see the font settings dialog box in the view -> application > settings -> font settings. The options I see are small, medium > and large font sizes and already small is selected. I am still > not able to see the "OK" window. Is there something I am missing? > > > Were the fonts always that big in Slicer3 for you? Is this something > that started recently? I assume the rest of the GUI must be pretty > much unusable because of this size. Are you using it on Linux, or > from a virtual machine like VMWare? > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From harish.slicer at gmail.com Wed Apr 15 13:48:17 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 15 Apr 2009 10:48:17 -0700 Subject: [slicer-devel] colors in linux In-Reply-To: <49E61BB9.7070904@bwh.harvard.edu> References: <20090415170000.487A753FD@public.kitware.com> <20090415170944.EB9644A8@public.kitware.com> <49E61BB9.7070904@bwh.harvard.edu> Message-ID: Hi Steve, This is my linux distribution Red Hat Enterprise Linux WS release 4 (Nahant) Thanks, Harish On 4/15/09, Steve Pieper wrote: > > Ron also saw this on one of his linux machines, but it doesn't seem to > occur on all machines. As far as Wendy or I know there was nothing in > slicer that changed to impact this. Perhaps there's a linux font package > that changed? > > Harish, what distribution do you use? I tested on ubuntu 8.10 and didn't > have a problem. > > -Steve > > > > > > Harish Doddi wrote: > >> The font is the same every time. The font seems ok for the rest of the >> GUI. I am using it on Linux and am using Slicer 3.3. Is there any feature >> related to this recently incorporated in slicer? >> >> Thanks, >> Harish >> >> On Wed, Apr 15, 2009 at 10:09 AM, Sebastien BARRE < >> sebastien.barre at kitware.com > wrote: >> >> At 4/15/2009 01:06 PM, Harish Doddi wrote: >> >> Hi Sebastien, >> >> I see the font settings dialog box in the view -> application >> settings -> font settings. The options I see are small, medium >> and large font sizes and already small is selected. I am still >> not able to see the "OK" window. Is there something I am missing? >> >> >> Were the fonts always that big in Slicer3 for you? Is this something >> that started recently? I assume the rest of the GUI must be pretty >> much unusable because of this size. Are you using it on Linux, or >> from a virtual machine like VMWare? >> >> >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090415/fcaa37c2/attachment-0001.html From sebastien.barre at kitware.com Wed Apr 15 14:02:00 2009 From: sebastien.barre at kitware.com (Sebastien BARRE) Date: Wed, 15 Apr 2009 14:02:00 -0400 Subject: [slicer-devel] colors in linux In-Reply-To: References: <20090415170000.487A753FD@public.kitware.com> <20090415170944.EB9644A8@public.kitware.com> <49E61BB9.7070904@bwh.harvard.edu> Message-ID: <20090415180229.60AF55590@public.kitware.com> At 4/15/2009 01:48 PM, Harish Doddi wrote: >Hi Steve, > >This is my linux distribution > >Red Hat Enterprise Linux WS release 4 (Nahant) Alright, I'm not sure what's going on here, but I've committed a workaround to both the KWWidgets HEAD and Slicer-3-4 branch. It's not great but should help you for now... Update your KWWidgets, build, and the color dialog can now be resized like a regular window: expand it to bring the OK button back in view. There is only one instance of this dialog used throughout the application, for speed, therefore you "only" have to perform this resizing trick once per Slicer3 session... From pieper at bwh.harvard.edu Wed Apr 15 18:57:22 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 15 Apr 2009 18:57:22 -0400 Subject: [slicer-devel] Slicer-3-4 branch is created - Re: update on bugs and release timing In-Reply-To: <49E4F948.4020503@bwh.harvard.edu> References: <49DBCF0E.6060606@bwh.harvard.edu> <49E4F948.4020503@bwh.harvard.edu> Message-ID: <49E66652.1080409@bwh.harvard.edu> New code can start going into the trunk now. Before long we'll also bump the trunk to VTK 5.4 and the latest ITK and other libs as needed. -Steve Steve Pieper wrote: > Update: after over 100 svn commits in the past week, we're now down to 8 > bugs that are tagged for fix before the release - most of these are > pretty small and/or just require confirmation before closing. Thanks, > developers, for all the hard work! > > So we will be creating the 3.4 release branch tomorrow afternoon and > start making release candidate builds. > > Please do not check in any changes unless they are on the bug list or > reported to the list. > > Thanks, > Steve > > Steve Pieper wrote: >> Hi - >> >> After going through the bug list we've identified 19 bugs that are >> tagged as "3.4 Targeted fix". The link should get you to them or you >> can see the attached image. >> >> http://tinyurl.com/cu9bzp >> >> Our plan now is to create the 3.4 release branch on Tuesday April 14. >> >> ** Any fixes after that date will need to be manually checked into both >> the trunk and the release branch. >> >> Please keep fixing bugs and testing the nightly builds. Add any new >> issues to the bug tracker and flag them if you think they are critical >> for the release. >> >> Thanks, >> Steve >> >> ------------------------------------------------------------------------ >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From harish.slicer at gmail.com Thu Apr 16 15:29:43 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Thu, 16 Apr 2009 12:29:43 -0700 Subject: [slicer-devel] Slicer Segmentation Violation Message-ID: Hi all, I just downloaded the source code of the new stable version of Slicer and build it in Linux. I see the binary Slicer3 in slicer-build directory. When I run it, I get this error [harishd at clark-del-un-7 Slicer3-build]$ ./Slicer3 -- version Starting Slicer: /home/harishd/RA/Stable_Slicer_Source_Code/Slicer3-build child killed: segmentation violation CHILDKILLED 19066 SIGSEGV {segmentation violation} and Slicer window exits. Is there some thing I am missing ? Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090416/2cd8676a/attachment.html From pieper at bwh.harvard.edu Thu Apr 16 15:42:00 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 16 Apr 2009 15:42:00 -0400 Subject: [slicer-devel] [slicer-users] Slicer Segmentation Violation In-Reply-To: References: Message-ID: <49E78A08.9070207@bwh.harvard.edu> Hi Harish - Coincidentally, we found out there's a bug in KWWidgets that is leading to a crash at startup. It's being actively worked on now and should be fixed soon. -Steve Harish Doddi wrote: > > > Hi all, > > I just downloaded the source code of the new stable version of Slicer > and build it in Linux. I see the binary Slicer3 in slicer-build > directory. When I run it, I get this error > > [harishd at clark-del-un-7 Slicer3-build]$ ./Slicer3 -- version > Starting Slicer: /home/harishd/RA/Stable_Slicer_Source_Code/Slicer3-build > > child killed: segmentation violation > > CHILDKILLED 19066 SIGSEGV {segmentation violation} > > and Slicer window exits. Is there some thing I am missing ? > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From sebastien.barre at kitware.com Fri Apr 17 11:41:05 2009 From: sebastien.barre at kitware.com (Sebastien BARRE) Date: Fri, 17 Apr 2009 11:41:05 -0400 Subject: [slicer-devel] [slicer-users] Slicer Segmentation Violation In-Reply-To: <49E78A08.9070207@bwh.harvard.edu> References: <49E78A08.9070207@bwh.harvard.edu> Message-ID: <20090417154109.B34E63822@public.kitware.com> Harish, At 4/16/2009 03:42 PM, Steve Pieper wrote: >Hi Harish - > >Coincidentally, we found out there's a bug in KWWidgets that is leading >to a crash at startup. It's being actively worked on now and should be >fixed soon. I fixed it yesterday evening, you should be good today. Thanks From sebastien.barre at kitware.com Fri Apr 17 12:18:26 2009 From: sebastien.barre at kitware.com (Sebastien BARRE) Date: Fri, 17 Apr 2009 12:18:26 -0400 Subject: [slicer-devel] [slicer-users] Slicer Segmentation Violation In-Reply-To: <49E78A08.9070207@bwh.harvard.edu> References: <49E78A08.9070207@bwh.harvard.edu> Message-ID: <20090417161830.657294DD0@public.kitware.com> Actually I'm seeing that segmentation fault at startup as well, and the KWW fix is in, so that's probably something else. I can't test on Win32 because the CUDA code does not compile for me: [ 85%] Building CXX object Modules/VolumeRendering/VolumeRenderingCuda/CMakeFiles/VolumeRenderingCuda.dir/vtkCudaVolumeMapperDummy.obj vtkCudaVolumeMapperDummy.cxx d:\src\kitware\slicer\slicer3-head\modules\volumerendering\volumerenderingcuda\v tkcudavolumemapperdummy.cxx(53) : error C4716: 'vtkCudaVolumeMapper::GetCurrentR enderMode' : must return a value make[2]: *** [Modules/VolumeRendering/VolumeRenderingCuda/CMakeFiles/VolumeRende ringCuda.dir/vtkCudaVolumeMapperDummy.obj] Error 2 make[1]: *** [Modules/VolumeRendering/VolumeRenderingCuda/CMakeFiles/VolumeRende ringCuda.dir/all] Error 2 make: *** [all] Error 2 Is the reposition still in feature-freeze mode for the release? If that's the case, should we put CUDA at the last moment? Seb At 4/16/2009 03:42 PM, Steve Pieper wrote: >Hi Harish - > >Coincidentally, we found out there's a bug in KWWidgets that is leading >to a crash at startup. It's being actively worked on now and should be >fixed soon. > >-Steve > >Harish Doddi wrote: > > > > > > Hi all, > > > > I just downloaded the source code of the new stable version of Slicer > > and build it in Linux. I see the binary Slicer3 in slicer-build > > directory. When I run it, I get this error > > > > [harishd at clark-del-un-7 Slicer3-build]$ ./Slicer3 -- version > > Starting Slicer: /home/harishd/RA/Stable_Slicer_Source_Code/Slicer3-build > > > > child killed: segmentation violation > > > > CHILDKILLED 19066 SIGSEGV {segmentation violation} > > > > and Slicer window exits. Is there some thing I am missing ? > > > > Thanks, > > Harish > > > > > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > slicer-users mailing list > > slicer-users at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > > To unsubscribe: send email to > slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject >_______________________________________________ >slicer-devel mailing list >slicer-devel at bwh.harvard.edu >http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >To unsubscribe: send email to >slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From harish.slicer at gmail.com Fri Apr 17 17:26:51 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Fri, 17 Apr 2009 14:26:51 -0700 Subject: [slicer-devel] EM Segmentation Problem Message-ID: Hi all, I am trying to apply EM Segmentation algorithm on knee dataset. I have created a filled label map of patella and am giving that as input atlas in the segmentation procedure. When I come to the last step of segmentation process, I get an error saying Scalar type mismatch for input images; all image scalar types must be the same (attached is a screenshot of it) . As of now I dont have any background atlas of knee so I dont give any background atlas in the input. I feel like I am missing something in giving correct input. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090417/0edeee46/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-Input Image Error.png Type: image/png Size: 19439 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090417/0edeee46/attachment-0001.png From harish.slicer at gmail.com Fri Apr 17 17:44:06 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Fri, 17 Apr 2009 14:44:06 -0700 Subject: [slicer-devel] EM Segmentation Problem In-Reply-To: References: Message-ID: Hi all, I even tried converting all the images into .nhdr and labelmap into .nhdr so that they are in the same format, but this gives the error message. Harish On 4/17/09, Harish Doddi wrote: > > > > Hi all, > > I am trying to apply EM Segmentation algorithm on knee dataset. I have > created a filled label map of patella and am giving that as input atlas in > the segmentation procedure. When I come to the last step of segmentation > process, I get an error saying Scalar type mismatch for input images; all > image scalar types must be the same (attached is a screenshot of it) . As of > now I dont have any background atlas of knee so I dont give any background > atlas in the input. I feel like I am missing something in giving correct > input. > > Thanks, > Harish > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090417/aa9098a0/attachment.html From nicole at bwh.harvard.edu Mon Apr 20 09:42:01 2009 From: nicole at bwh.harvard.edu (Nicole Aucoin) Date: Mon, 20 Apr 2009 09:42:01 -0400 (EDT) Subject: [slicer-devel] EM Segmentation Problem In-Reply-To: References: Message-ID: Hi Harish, The scalar type is related to the values that can be held in each voxel rather than the file format. If you go into the Volumes module, there's an info tab near the bottom that you can expand and it should tell you if your volume scalar types are short, unsigned char, float etc. I'm pretty sure that there's an image cast filter available so that you can cast all of your images into the same scalar type, but I'm currently rebuilding my copy of Slicer and can't point it out right now. Nicole On Fri, 17 Apr 2009, Harish Doddi wrote: > Hi all, > > I even tried converting all the images into .nhdr and labelmap into .nhdr so > that they are in the same format, but this gives the error message. > > Harish > > On 4/17/09, Harish Doddi wrote: >> >> >> >> Hi all, >> >> I am trying to apply EM Segmentation algorithm on knee dataset. I have >> created a filled label map of patella and am giving that as input atlas in >> the segmentation procedure. When I come to the last step of segmentation >> process, I get an error saying Scalar type mismatch for input images; all >> image scalar types must be the same (attached is a screenshot of it) . As of >> now I dont have any background atlas of knee so I dont give any background >> atlas in the input. I feel like I am missing something in giving correct >> input. >> >> Thanks, >> Harish >> >> > -- Nicole Aucoin nicole at bwh.harvard.edu Surgical Planning Laboratory, Brigham and Women's Hospital Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 http://www.slicer.org From harish.slicer at gmail.com Mon Apr 20 13:50:08 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 20 Apr 2009 10:50:08 -0700 Subject: [slicer-devel] EM Segmentation Problem In-Reply-To: References: Message-ID: Hi Nicole, Yes you are correct. One of volumes is in short and label map(atlas) is in unsigned char. Could someone tell which module does this casting ? Thanks, Harish On 4/20/09, Nicole Aucoin wrote: > > Hi Harish, > > The scalar type is related to the values that can be held in each voxel > rather than the file format. If you go into the Volumes module, there's an > info tab near the bottom that you can expand and it should tell you if your > volume scalar types are short, unsigned char, float etc. > I'm pretty sure that there's an image cast filter available so that you can > cast all of your images into the same scalar type, but I'm currently > rebuilding my copy of Slicer and can't point it out right now. > > Nicole > > On Fri, 17 Apr 2009, Harish Doddi wrote: > > Hi all, >> >> I even tried converting all the images into .nhdr and labelmap into .nhdr >> so >> that they are in the same format, but this gives the error message. >> >> Harish >> >> On 4/17/09, Harish Doddi wrote: >> >>> >>> >>> >>> Hi all, >>> >>> I am trying to apply EM Segmentation algorithm on knee dataset. I have >>> created a filled label map of patella and am giving that as input atlas >>> in >>> the segmentation procedure. When I come to the last step of segmentation >>> process, I get an error saying Scalar type mismatch for input images; all >>> image scalar types must be the same (attached is a screenshot of it) . As >>> of >>> now I dont have any background atlas of knee so I dont give any >>> background >>> atlas in the input. I feel like I am missing something in giving correct >>> input. >>> >>> Thanks, >>> Harish >>> >>> >>> >> > -- > Nicole Aucoin nicole at bwh.harvard.edu > Surgical Planning Laboratory, Brigham and Women's Hospital > Rm 206, 1249 Boylston St., Boston, MA 02215 617-525-6216 > http://www.slicer.org > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090420/7cabfbab/attachment.html From hans-johnson at uiowa.edu Tue Apr 21 22:10:15 2009 From: hans-johnson at uiowa.edu (Hans Johnson) Date: Tue, 21 Apr 2009 21:10:15 -0500 Subject: [slicer-devel] LibProstateNav.dylib causes segmentation fault Message-ID: When I attempt to start slicer I get a segmentation fault during startup. Once I remove the LibProstateNav.dylib file, slicer works again. Hans Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged. If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited. Please reply to the sender that you have received the message in error, then delete it. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090421/278c2a0d/attachment.html From hans-johnson at uiowa.edu Tue Apr 21 22:14:46 2009 From: hans-johnson at uiowa.edu (Hans Johnson) Date: Tue, 21 Apr 2009 21:14:46 -0500 Subject: [slicer-devel] Recent default change? Message-ID: With recent builds of slicer, all my images load by default to look like: After adjusting the display, they look OK again. The Auto Window level does not seem to be doing a very good job: Thanks, Hans Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged. If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited. Please reply to the sender that you have received the message in error, then delete it. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090421/83877c6a/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 64716 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090421/83877c6a/attachment-0002.png -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 1798 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090421/83877c6a/attachment-0003.png From hliu at bwh.harvard.edu Tue Apr 21 22:17:04 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Tue, 21 Apr 2009 22:17:04 -0400 Subject: [slicer-devel] LibProstateNav.dylib causes segmentation fault In-Reply-To: References: Message-ID: <49EE7E20.5030104@bwh.harvard.edu> Yes, please comment out this module when compiling Slicer3. We're working on this issue. Thanks. Haiying Hans Johnson wrote: > > When I attempt to start slicer I get a segmentation fault during startup. > > Once I remove the LibProstateNav.dylib file, slicer works again. > > Hans > > Notice: This UI Health Care e-mail (including attachments) is covered > by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is > confidential and may be legally privileged. If you are not the > intended recipient, you are hereby notified that any retention, > dissemination, distribution, or copying of this communication is > strictly prohibited. Please reply to the sender that you have > received the message in error, then delete it. Thank you. > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Wed Apr 22 09:02:43 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 22 Apr 2009 09:02:43 -0400 Subject: [slicer-devel] LibProstateNav.dylib causes segmentation fault In-Reply-To: <49EE7E20.5030104@bwh.harvard.edu> References: <49EE7E20.5030104@bwh.harvard.edu> Message-ID: <49EF1573.1030009@bwh.harvard.edu> Also, since the module is disabled from the CMakeLists.txt file, it doesn't know that the old shared libraries are out of date - this is why you may need to manually remove the old shared libraries. -Steve Haiying Liu wrote: > Yes, please comment out this module when compiling Slicer3. We're > working on this issue. Thanks. Haiying > > Hans Johnson wrote: >> When I attempt to start slicer I get a segmentation fault during startup. >> >> Once I remove the LibProstateNav.dylib file, slicer works again. >> >> Hans >> >> Notice: This UI Health Care e-mail (including attachments) is covered >> by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is >> confidential and may be legally privileged. If you are not the >> intended recipient, you are hereby notified that any retention, >> dissemination, distribution, or copying of this communication is >> strictly prohibited. Please reply to the sender that you have >> received the message in error, then delete it. Thank you. >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Wed Apr 22 09:08:46 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 22 Apr 2009 09:08:46 -0400 Subject: [slicer-devel] Recent default change? In-Reply-To: References: Message-ID: <49EF16DE.8010603@bwh.harvard.edu> I'm not seeing a change in that behavior -- can you send some anonymized example data? The algorithm slicer uses, imperfect to be sure, assume there the image consists of a dark background (air) and some non-dark tissue; it tries to pick a window/level based on the non-dark image data. It works pretty well for MR, but I have noticed that after some processing steps (filters) it can guess badly. Is that the case here or are they 'fresh' MR scans? -Steve Hans Johnson wrote: > With recent builds of slicer, all my images load by default to look like: > > > > After adjusting the display, they look OK again. > > The Auto Window level does not seem to be doing a very good job: > > > > Thanks, > Hans > > Notice: This UI Health Care e-mail (including attachments) is covered by > the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is > confidential and may be legally privileged. If you are not the intended > recipient, you are hereby notified that any retention, dissemination, > distribution, or copying of this communication is strictly prohibited. > Please reply to the sender that you have received the message in error, > then delete it. Thank you. > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From hans-johnson at uiowa.edu Wed Apr 22 11:10:21 2009 From: hans-johnson at uiowa.edu (Hans Johnson) Date: Wed, 22 Apr 2009 10:10:21 -0500 Subject: [slicer-devel] Recent default change? In-Reply-To: <49EF16DE.8010603@bwh.harvard.edu> Message-ID: Steve, They have been bias field corrected with itkEMS. I've de-identified 2 data sets that are ugly, but show the problem. Hans On 4/22/09 8:08 AM, "Steve Pieper" wrote: > I'm not seeing a change in that behavior -- can you send some anonymized > example data? The algorithm slicer uses, imperfect to be sure, assume > there the image consists of a dark background (air) and some non-dark > tissue; it tries to pick a window/level based on the non-dark image > data. It works pretty well for MR, but I have noticed that after some > processing steps (filters) it can guess badly. Is that the case here or > are they 'fresh' MR scans? > > -Steve > > Hans Johnson wrote: >> With recent builds of slicer, all my images load by default to look like: >> >> >> >> After adjusting the display, they look OK again. >> >> The Auto Window level does not seem to be doing a very good job: >> >> >> >> Thanks, >> Hans >> >> Notice: This UI Health Care e-mail (including attachments) is covered by >> the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is >> confidential and may be legally privileged. If you are not the intended >> recipient, you are hereby notified that any retention, dissemination, >> distribution, or copying of this communication is strictly prohibited. >> Please reply to the sender that you have received the message in error, >> then delete it. Thank you. >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject From pieper at bwh.harvard.edu Wed Apr 22 12:08:45 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 22 Apr 2009 12:08:45 -0400 Subject: [slicer-devel] Recent default change? In-Reply-To: References: Message-ID: <49EF410D.6030303@bwh.harvard.edu> Okay, great - can we put them in a bug report - my guess is that they don't match the assumptions built into the current auto w/l guesser. Probably we need several different algorithms rather than a one-size-fits all auto mode. -Steve Hans Johnson wrote: > Steve, > > They have been bias field corrected with itkEMS. > > I've de-identified 2 data sets that are ugly, but show the problem. > > Hans > > > On 4/22/09 8:08 AM, "Steve Pieper" wrote: > >> I'm not seeing a change in that behavior -- can you send some anonymized >> example data? The algorithm slicer uses, imperfect to be sure, assume >> there the image consists of a dark background (air) and some non-dark >> tissue; it tries to pick a window/level based on the non-dark image >> data. It works pretty well for MR, but I have noticed that after some >> processing steps (filters) it can guess badly. Is that the case here or >> are they 'fresh' MR scans? >> >> -Steve >> >> Hans Johnson wrote: >>> With recent builds of slicer, all my images load by default to look like: >>> >>> >>> >>> After adjusting the display, they look OK again. >>> >>> The Auto Window level does not seem to be doing a very good job: >>> >>> >>> >>> Thanks, >>> Hans >>> >>> Notice: This UI Health Care e-mail (including attachments) is covered by >>> the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is >>> confidential and may be legally privileged. If you are not the intended >>> recipient, you are hereby notified that any retention, dissemination, >>> distribution, or copying of this communication is strictly prohibited. >>> Please reply to the sender that you have received the message in error, >>> then delete it. Thank you. >>> >>> >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject > From fedorov at bwh.harvard.edu Wed Apr 22 12:18:21 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Wed, 22 Apr 2009 12:18:21 -0400 Subject: [slicer-devel] Recent default change? In-Reply-To: <49EF410D.6030303@bwh.harvard.edu> References: <49EF410D.6030303@bwh.harvard.edu> Message-ID: <82f536810904220918s42ff41e7jc4c8768fd7f8d4c8@mail.gmail.com> Steve, I also observed similar problems on the un-processed MR data. Try to load the meningioma image from the Slicer test data in the trunk here: Testing/Data/Input/MRMeningioma1.nrrd I think the auto setting is not particularly good for this image too. Andriy Fedorov On Wed, Apr 22, 2009 at 12:08 PM, Steve Pieper wrote: > Okay, great - can we put them in a bug report - my guess is that they > don't match the assumptions built into the current auto w/l guesser. > Probably we need several different algorithms rather than a > one-size-fits all auto mode. > > -Steve > > Hans Johnson wrote: >> Steve, >> >> They have been bias field corrected with itkEMS. >> >> I've de-identified 2 data sets that are ugly, but show the problem. >> >> Hans >> >> >> On 4/22/09 8:08 AM, "Steve Pieper" wrote: >> >>> I'm not seeing a change in that behavior -- can you send some anonymized >>> example data? ?The algorithm slicer uses, imperfect to be sure, assume >>> there the image consists of a dark background (air) and some non-dark >>> tissue; it tries to pick a window/level based on the non-dark image >>> data. ?It works pretty well for MR, but I have noticed that after some >>> processing steps (filters) it can guess badly. ?Is that the case here or >>> are they 'fresh' MR scans? >>> >>> -Steve >>> >>> Hans Johnson wrote: >>>> With recent builds of slicer, all my images load by default to look like: >>>> >>>> >>>> >>>> After adjusting the display, they look OK again. >>>> >>>> The Auto Window level does not seem to be doing a very good job: >>>> >>>> >>>> >>>> Thanks, >>>> Hans >>>> >>>> Notice: This UI Health Care e-mail (including attachments) is covered by >>>> the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is >>>> confidential and may be legally privileged. ?If you are not the intended >>>> recipient, you are hereby notified that any retention, dissemination, >>>> distribution, or copying of this communication is strictly prohibited. >>>> Please reply to the sender that you have received the message in error, >>>> then delete it. ?Thank you. >>>> >>>> >>>> ------------------------------------------------------------------------ >>>> >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>>> with unsubscribe as the subject >> > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From enjoywm at cs.wm.edu Thu Apr 23 12:10:21 2009 From: enjoywm at cs.wm.edu (Yixun Liu) Date: Thu, 23 Apr 2009 12:10:21 -0400 Subject: [slicer-devel] Slicer support color image segmentation? Message-ID: <49F092ED.9000601@cs.wm.edu> Hi, Does Slicer support the segmentation of the color or RGB image? Thanks. Yixun From enjoywm at cs.wm.edu Thu Apr 23 12:16:45 2009 From: enjoywm at cs.wm.edu (Yixun Liu) Date: Thu, 23 Apr 2009 12:16:45 -0400 Subject: [slicer-devel] color image segmentation Message-ID: <49F0946D.70704@cs.wm.edu> Hi, Does Slicer support the segmentation of the color or RGB image? Thanks. Yixun From pieper at bwh.harvard.edu Thu Apr 23 13:12:24 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 23 Apr 2009 13:12:24 -0400 Subject: [slicer-devel] color image segmentation In-Reply-To: <49F0946D.70704@cs.wm.edu> References: <49F0946D.70704@cs.wm.edu> Message-ID: <49F0A178.9000500@bwh.harvard.edu> Not currently, no. We hope to add that in the future. -Steve Yixun Liu wrote: > Hi, > Does Slicer support the segmentation of the color or RGB image? > > Thanks. > > Yixun > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From wschramm at bwh.harvard.edu Thu Apr 23 14:10:22 2009 From: wschramm at bwh.harvard.edu (Wolfgang Schramm) Date: Thu, 23 Apr 2009 14:10:22 -0400 Subject: [slicer-devel] =?utf-8?q?CmakeLists_Problem_under_Windows=3F?= Message-ID: <1099f91f754451f51af7d12e2433d5a6@bwh.harvard.edu> Hello, since this morning I got the following error when building Slicer (Trunk, Python turned off) under Visual Studio 2008 32Bit (Windows XP SP3): CMake Error: Error in cmake code at C:/cygwin/home/Wolfgang/Slicer3/Applications/CLI/CMakeLists.txt:584: Parse error. Function missing ending ")". Instead found left paren with text "(". Line 584 is: if ( ${ITK_VERSION_MAJOR} GREATER 3 OR (${ITK_VERSION_MAJOR} MATCHES 3 AND ${ITK_VERSION_MINOR} GREATER 12) ) Also the lines above look ok, can't find an obvious error. Also it seems to build fine under Linux. Can anybody confirm this problem? Thank you Wolfgang From pieper at bwh.harvard.edu Thu Apr 23 14:34:37 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 23 Apr 2009 14:34:37 -0400 Subject: [slicer-devel] CmakeLists Problem under Windows? In-Reply-To: <1099f91f754451f51af7d12e2433d5a6@bwh.harvard.edu> References: <1099f91f754451f51af7d12e2433d5a6@bwh.harvard.edu> Message-ID: <49F0B4BD.9010000@bwh.harvard.edu> Interesting - yes, I see. I just checked in a fix by rearranging the logic. It looks like the CMake 2.6.0 binary we use with windows doesn't handle the nested boolean logic in the if statement. I will also put this in the release branch. -Steve Wolfgang Schramm wrote: > Hello, > > since this morning I got the following error when building Slicer (Trunk, > Python turned off) under Visual Studio 2008 32Bit (Windows XP SP3): > > CMake Error: Error in cmake code at > C:/cygwin/home/Wolfgang/Slicer3/Applications/CLI/CMakeLists.txt:584: > Parse error. Function missing ending ")". Instead found left paren with > text "(". > > Line 584 is: if ( ${ITK_VERSION_MAJOR} GREATER 3 OR (${ITK_VERSION_MAJOR} > MATCHES 3 AND ${ITK_VERSION_MINOR} GREATER 12) ) > > Also the lines above look ok, can't find an obvious error. Also it seems to > build fine under Linux. > > Can anybody confirm this problem? > > Thank you > Wolfgang > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Fri Apr 24 12:21:02 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 24 Apr 2009 12:21:02 -0400 Subject: [slicer-devel] FYI - updated build scripts Message-ID: <49F1E6EE.1040504@bwh.harvard.edu> Hi - Attila Nagy has kindly provided patches for to build slicer3 on solaris in 32 and 64 bit modes. These shouldn't impact other platforms, but speak up if anything starts breaking. I just checked these into the trunk: http://viewvc.slicer.org/viewcvs.cgi?rev=9316&view=rev Thanks, -Steve From harish.slicer at gmail.com Fri Apr 24 15:04:54 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Fri, 24 Apr 2009 12:04:54 -0700 Subject: [slicer-devel] Register Images Message-ID: Hi all, I am trying to use the module Register Images in slicer. Here I see that we can give Load Transform as input. Can the load transform input file be generated by some other module given two image volumes or it should be generated manually ? Thanks, harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090424/d3306206/attachment.html From kikinis at bwh.harvard.edu Fri Apr 24 15:31:50 2009 From: kikinis at bwh.harvard.edu (Ron Kikinis) Date: Fri, 24 Apr 2009 15:31:50 -0400 Subject: [slicer-devel] Register Images In-Reply-To: References: Message-ID: <49F213A6.4000106@bwh.harvard.edu> Hi, go to the help tab on that module and click on the link to the slicer wiki. there should be some discussion about this. The module is not yet fully integrated into Slicer. Depending on your needs, you might want to look at the affine and b-spline modules as well. Ron Harish Doddi wrote: > > > Hi all, > > I am trying to use the module Register Images in slicer. Here I see that > we can give Load Transform as input. Can the load transform input file > be generated by some other module given two image volumes or it should > be generated manually ? > > Thanks, > harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject -- Ron Kikinis, M.D., Professor of Radiology, Harvard Medical School Director, Surgical Planning Laboratory http://www.spl.harvard.edu/~kikinis From simon at math.uh.edu Fri Apr 24 16:56:05 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Fri, 24 Apr 2009 15:56:05 -0500 Subject: [slicer-devel] creating a new scalar volume and setting it up correctly. Message-ID: <200904241556.05902.simon@math.uh.edu> Hi, Can someone point me at the best place in the docs and/or code to see how to properly introduce a new scalar data volume? I'm loading some data from an (unsupported, non-standard) file format and while I have it mostly correct, there are odd things going on with the volume suggesting that I've missed some needed setup. For example, if I turn interpolate off and then on from the volume panel, the window and level adjusts itself. While I'm at it, is there a way to set/change the default file format that these volumes will want to save as, or do I have to go and change each one? Thanks, Simon. From pieper at bwh.harvard.edu Mon Apr 27 08:19:05 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 27 Apr 2009 08:19:05 -0400 Subject: [slicer-devel] creating a new scalar volume and setting it up correctly. In-Reply-To: <200904241556.05902.simon@math.uh.edu> References: <200904241556.05902.simon@math.uh.edu> Message-ID: <49F5A2B9.3080305@bwh.harvard.edu> Hi Simon - Probably the most general solution would be to create an ITK IO factory plugin as described here: http://www.itk.org/Wiki/Plugin_IO_mechanisms There are several good examples you could follow. Such a factory would be discovered by slicer at run time and should "just work" with the rest of the system. Or, if you are working programmatically inside slicer you could look at this method to see how to set up the data structures: vtkMRMLScalarVolumeNode* vtkSlicerVolumesLogic:: CloneVolume (vtkMRMLScene *scene, vtkMRMLVolumeNode *volumeNode, const char *name) There's no user-selectable way to set the default file format now -- that would make a good feature request for the bug tracker. Best, Steve Simon K Alexander wrote: > Hi, > > Can someone point me at the best place in the docs and/or code to see how to properly introduce a new scalar data > volume? I'm loading some data from an (unsupported, non-standard) file format and while I have it mostly correct, > there are odd things going on with the volume suggesting that I've missed some needed setup. For example, if I turn > interpolate off and then on from the volume panel, the window and level adjusts itself. > > While I'm at it, is there a way to set/change the default file format that these volumes will want to save as, or do I > have to go and change each one? > > Thanks, > Simon. > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From simon at math.uh.edu Mon Apr 27 15:53:14 2009 From: simon at math.uh.edu (Simon K Alexander) Date: Mon, 27 Apr 2009 14:53:14 -0500 Subject: [slicer-devel] creating a new scalar volume and setting it up correctly. In-Reply-To: <49F5A2B9.3080305@bwh.harvard.edu> References: <200904241556.05902.simon@math.uh.edu> <49F5A2B9.3080305@bwh.harvard.edu> Message-ID: <200904271453.14561.simon@math.uh.edu> Hi Steve, Thanks for your suggestions. I had a brief look at the ITK IO factory idea but somewhat rejected it. I probably should have mentioned that my non-standard format is contained in a dicom structure, but has unusual tags. At the moment I'm just parsing the dicom structure myself. I suspect making this work properly with the more general dicom reader is more work. Supporting this won't be any use to anyone else, either. The other thing about this data is that it contains a time series of volumes, so any reader need to allow me to generate multiple volumes in a particular way. These two problems combined had led me away from the idea of hooking into the regular I/O mechanism, but I may give that a second look. I'll have a look at the logic in Clone Volume, thanks. On Monday 27 April 2009 07:19:05 am Steve Pieper wrote: > Hi Simon - > > Probably the most general solution would be to create an ITK IO factory > plugin as described here: > > http://www.itk.org/Wiki/Plugin_IO_mechanisms > > There are several good examples you could follow. Such a factory would > be discovered by slicer at run time and should "just work" with the rest > of the system. > > Or, if you are working programmatically inside slicer you could look at > this method to see how to set up the data structures: > > vtkMRMLScalarVolumeNode* > vtkSlicerVolumesLogic:: > CloneVolume (vtkMRMLScene *scene, > vtkMRMLVolumeNode *volumeNode, > const char *name) > > There's no user-selectable way to set the default file format now -- > that would make a good feature request for the bug tracker. > > Best, > Steve > > Simon K Alexander wrote: > > Hi, > > > > Can someone point me at the best place in the docs and/or code to see how > > to properly introduce a new scalar data volume? I'm loading some data > > from an (unsupported, non-standard) file format and while I have it > > mostly correct, there are odd things going on with the volume suggesting > > that I've missed some needed setup. For example, if I turn interpolate > > off and then on from the volume panel, the window and level adjusts > > itself. > > > > While I'm at it, is there a way to set/change the default file format > > that these volumes will want to save as, or do I have to go and change > > each one? > > > > Thanks, > > Simon. > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > > subject -- Simon Alexander, Ph.D Department of Mathematics University of Houston 619 PGH phone: 713-743-0220 fax: 713-743-3505 From harish.slicer at gmail.com Wed Apr 29 19:45:23 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 29 Apr 2009 16:45:23 -0700 Subject: [slicer-devel] Stanford Simbios Group Message-ID: Hi all, Saikat and me have been working on the knee segmentation project. As part of it, we have built our project wiki page which contains the datasets we use and the progress we have done so far. The url is http://slicer.spl.harvard.edu/slicerWiki/index.php/Stanford_Simbios_group You can visit this page and take our test datasets to experiment on. We are happy to answer any questions you might have. Any feedback is welcome. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090429/d3f3c5dd/attachment.html From nrannou at bwh.harvard.edu Fri May 1 21:29:59 2009 From: nrannou at bwh.harvard.edu (Nicolas Rannou) Date: Fri, 01 May 2009 21:29:59 -0400 Subject: [slicer-devel] Wrapping application/method in TCL Message-ID: <49FBA217.1080608@bwh.harvard.edu> Hi all, I'm currently developping an application for Slicer but errors occurs, as if I didn't wrapped my application in TCL. I have an issue trying to use vtkKWParameterValueFunctionEditor::SetParameterCursorMovingCommand(vtkObject * object,const char * method). I added a cursor in my histogram and I would like to call a function each time I move the cursor. Which should be possible with the function above. So I created a function 'GetValue()' (to try). This function does not do anything yet. I called this function like this: /*******************************************************/ ...-> SetParameterCursorMovingCommand( NULL, "GetValue"); /*******************************************************/ The problem is that each time I move my cursor, an error occurs: /*******************************************************/ ERROR: In /projects/nacvisitors/nrannou/Slicer3/Slicer3-lib/KWWidgets/vtkKWTkUtilities.cxx, line 230 vtkSlicerApplication (0xefa5b0): Script: GetValue Returned Error on line 1: invalid command name "GetValue" Stack trace: invalid command name "GetValue" while executing "GetValue" ERROR: In /projects/nacvisitors/nrannou/Slicer3/Slicer3-lib/KWWidgets/vtkKWTkUtilities.cxx, line 230 vtkSlicerApplication (0xefa5b0): Script: GetValue 960.666667 Returned Error on line 1: invalid command name "GetValue" Stack trace: invalid command name "GetValue" while executing "GetValue 960.666667" /*******************************************************/ It seems like it is not wrapped. The point is that I wrapped my application using in the CMakeList.txt: /*******************************************/ include("${VTK_CMAKE_DIR}/vtkWrapTcl.cmake") vtk_wrap_tcl3(EMSegment EMSegment_TCL_SRCS "${EMSegment_SRCS}" "") /*******************************************/ And I couldn't find any documentation which helps about that. Thanks, Nicolas From pieper at bwh.harvard.edu Sat May 2 10:48:35 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Sat, 02 May 2009 10:48:35 -0400 Subject: [slicer-devel] Wrapping application/method in TCL In-Reply-To: <49FBA217.1080608@bwh.harvard.edu> References: <49FBA217.1080608@bwh.harvard.edu> Message-ID: <49FC5D43.5060808@bwh.harvard.edu> Hi Nicolas - When you set a command callback, you need to specify which object instance on which you want to invoke the method. You are passing NULL: > /*******************************************************/ > ...-> SetParameterCursorMovingCommand( NULL, "GetValue"); > /*******************************************************/ You probably want something like: ...->SetParameterCursorMovingCommand( this, "GetValue"); -Steve Nicolas Rannou wrote: > Hi all, > > I'm currently developping an application for Slicer but errors occurs, > as if I didn't wrapped my application in TCL. > > I have an issue trying to use > vtkKWParameterValueFunctionEditor::SetParameterCursorMovingCommand(vtkObject > * object,const char * method). > I added a cursor in my histogram and I would like to call a function > each time I move the cursor. Which should be possible with the function > above. > So I created a function 'GetValue()' (to try). This function does not do > anything yet. > I called this function like this: > > /*******************************************************/ > ...-> SetParameterCursorMovingCommand( NULL, "GetValue"); > /*******************************************************/ > > The problem is that each time I move my cursor, an error occurs: > > /*******************************************************/ > ERROR: In > /projects/nacvisitors/nrannou/Slicer3/Slicer3-lib/KWWidgets/vtkKWTkUtilities.cxx, > line 230 > vtkSlicerApplication (0xefa5b0): > Script: > GetValue > Returned Error on line 1: > invalid command name "GetValue" > Stack trace: > invalid command name "GetValue" > while executing > "GetValue" > > ERROR: In > /projects/nacvisitors/nrannou/Slicer3/Slicer3-lib/KWWidgets/vtkKWTkUtilities.cxx, > line 230 > vtkSlicerApplication (0xefa5b0): > Script: > GetValue 960.666667 > Returned Error on line 1: > invalid command name "GetValue" > Stack trace: > invalid command name "GetValue" > while executing > "GetValue 960.666667" > /*******************************************************/ > > It seems like it is not wrapped. > The point is that I wrapped my application using in the CMakeList.txt: > > /*******************************************/ > include("${VTK_CMAKE_DIR}/vtkWrapTcl.cmake") > vtk_wrap_tcl3(EMSegment > EMSegment_TCL_SRCS > "${EMSegment_SRCS}" "") > /*******************************************/ > > And I couldn't find any documentation which helps about that. > > Thanks, > Nicolas > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From ralph.brecheisen at gmail.com Mon May 4 07:22:22 2009 From: ralph.brecheisen at gmail.com (Ralph Brecheisen) Date: Mon, 4 May 2009 13:22:22 +0200 Subject: [slicer-devel] Problen building Slicer3 for Visual Studio 2008 Message-ID: <368a82280905040422h9e15ffdm6a69045e26e1e0f8@mail.gmail.com> Hi, I'm following the build instructions for building Slicer3 on Windows Vista and Visual Studio 2008 Express Edition. However, the "getbuildtest.tcl" script I run from Cygwin at some point tries to execute the following line: "running : {c:/Program Files/Microsoft Visual Studio 9.0/Common7/IDE/VCExpress.exe} PCBuild/pcbuild.sln /Upgrade" Visual Studio reports that it doesn't know the switch '/Upgrade' en exits. The remainder of "getbuildtest.tcl" then also fails. Does anyone have experience with this problem as well? Thanks, Ralph -- Ir. Ralph Brecheisen Biomedical Image Analysis @ Technical University of Eindhoven, Den Dolech 2, 5600 MB, Eindhoven, Netherlands Image-Guided Neurosurgery @ Maastricht University Hospital P. Debyelaan 25, 6229 HX, Maastricht, Netherlands Tel: +31-40-2475571 (Eindhoven) Tel: +31-43-3876292 (Maastricht) E-mail: r.brecheisen at tue.nl, ralph.brecheisen at gmail.com -------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090504/afc94bd3/attachment.html From fedorov at bwh.harvard.edu Mon May 4 08:15:35 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Mon, 4 May 2009 08:15:35 -0400 Subject: [slicer-devel] Problen building Slicer3 for Visual Studio 2008 In-Reply-To: <368a82280905040422h9e15ffdm6a69045e26e1e0f8@mail.gmail.com> References: <368a82280905040422h9e15ffdm6a69045e26e1e0f8@mail.gmail.com> Message-ID: <82f536810905040515l6c78e9e2o2d7713b3e85164c4@mail.gmail.com> Ralph, Yes, this is a known issue. http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002882.html Andriy Fedorov On Mon, May 4, 2009 at 7:22 AM, Ralph Brecheisen wrote: > Hi, > > I'm following the build instructions for building Slicer3 on Windows Vista > and Visual Studio 2008 Express Edition. However, the "getbuildtest.tcl" > script I run from Cygwin at some point tries to execute the following line: > > "running : {c:/Program Files/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpress.exe} PCBuild/pcbuild.sln /Upgrade" > > Visual Studio reports that it doesn't know the switch '/Upgrade' en exits. > The remainder of "getbuildtest.tcl" then also fails. > > Does anyone have experience with this problem as well? > > Thanks, > > Ralph > > > > > -- > Ir. Ralph Brecheisen > > Biomedical Image Analysis @ Technical University of Eindhoven, Den Dolech 2, > 5600 MB, Eindhoven, Netherlands > Image-Guided Neurosurgery @ Maastricht University Hospital > P. Debyelaan 25, 6229 HX, Maastricht, Netherlands > > Tel: +31-40-2475571 (Eindhoven) > Tel: +31-43-3876292 (Maastricht) > E-mail: r.brecheisen at tue.nl, ralph.brecheisen at gmail.com > -------------------------------------------------------------------------------------------------- > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > From ralph.brecheisen at gmail.com Mon May 4 08:54:19 2009 From: ralph.brecheisen at gmail.com (Ralph Brecheisen) Date: Mon, 4 May 2009 14:54:19 +0200 Subject: [slicer-devel] Problen building Slicer3 for Visual Studio 2008 In-Reply-To: <82f536810905040515l6c78e9e2o2d7713b3e85164c4@mail.gmail.com> References: <368a82280905040422h9e15ffdm6a69045e26e1e0f8@mail.gmail.com> <82f536810905040515l6c78e9e2o2d7713b3e85164c4@mail.gmail.com> Message-ID: <368a82280905040554l6ee4dcbdl9dc3f75aebdf6ef0@mail.gmail.com> Hi Andriy, Thanks for the tip. I''ll try to switch Python off. Greetz, Rlp 2009/5/4 Andriy Fedorov > Ralph, > > Yes, this is a known issue. > > http://massmail.spl.harvard.edu/pipermail/slicer-devel/2009/002882.html > > Andriy Fedorov > > > > On Mon, May 4, 2009 at 7:22 AM, Ralph Brecheisen > wrote: > > Hi, > > > > I'm following the build instructions for building Slicer3 on Windows > Vista > > and Visual Studio 2008 Express Edition. However, the "getbuildtest.tcl" > > script I run from Cygwin at some point tries to execute the following > line: > > > > "running : {c:/Program Files/Microsoft Visual Studio > > 9.0/Common7/IDE/VCExpress.exe} PCBuild/pcbuild.sln /Upgrade" > > > > Visual Studio reports that it doesn't know the switch '/Upgrade' en > exits. > > The remainder of "getbuildtest.tcl" then also fails. > > > > Does anyone have experience with this problem as well? > > > > Thanks, > > > > Ralph > > > > > > > > > > -- > > Ir. Ralph Brecheisen > > > > Biomedical Image Analysis @ Technical University of Eindhoven, Den Dolech > 2, > > 5600 MB, Eindhoven, Netherlands > > Image-Guided Neurosurgery @ Maastricht University Hospital > > P. Debyelaan 25, 6229 HX, Maastricht, Netherlands > > > > Tel: +31-40-2475571 (Eindhoven) > > Tel: +31-43-3876292 (Maastricht) > > E-mail: r.brecheisen at tue.nl, ralph.brecheisen at gmail.com > > > -------------------------------------------------------------------------------------------------- > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu > > with unsubscribe as the subject > > > -- "I respectfully decline the invitation to join your hallucination." - Scott Adams -- Ir. Ralph Brecheisen Biomedical Image Analysis @ Technical University of Eindhoven, Den Dolech 2, 5600 MB, Eindhoven, Netherlands Image-Guided Neurosurgery @ Maastricht University Hospital P. Debyelaan 25, 6229 HX, Maastricht, Netherlands Tel: +31-40-2475571 (Eindhoven) Tel: +31-43-3876292 (Maastricht) E-mail: r.brecheisen at tue.nl, ralph.brecheisen at gmail.com -------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090504/3249f27d/attachment.html From norman-k-williams at uiowa.edu Mon May 4 10:35:52 2009 From: norman-k-williams at uiowa.edu (kent williams) Date: Mon, 04 May 2009 09:35:52 -0500 Subject: [slicer-devel] Wrapping application/method in TCL In-Reply-To: <49FBA217.1080608@bwh.harvard.edu> Message-ID: You have my sympathy. VTK's wrapping mechanism is pretty much entirely undocumented. I just went look with google and most of the hits on 'VTK Wrapping' point to my posts to the VTK users list about how wrapping was driving me insane. Nothing will get wrapped in your application unless you explicitly wrap it. Nearly everyone starts with the example of the MedicalImageViewer -- so much so I actually have a template application source directory that I start all my KWWidgets projects from, based on the MedicalImageViewer CMakeLists.txt files. To get a library wrapped, you work it this way: SET(LIB_NAME mylibrary) SET(LIB_SRCS a.cxx b.cxx c.xx) KWWidgets_WRAP_TCL(${LIB_NAME} LIB_TCL_SOURCES "${LIB_SRCS}" "") ADD_LIBRARY(${LIB_NAME} STATIC ${LIB_TCL_SRCS} ${LIB_SRCS}) Then link ${LIB_NAME} into your program. If you're not already doing this wrapping, you'll run into all sorts of compile errors because VTK wrapping isn't a full C++ parser. Anything too 'classy' will confuse it. The way around this is to surround stuff that won't matter for the purposes of wrapping with //BTX ...code that blows VTK Wrapping's mind //ETX Your class NEEDS to have a filename and class name beginning in 'vtk' -- otherwise wrapping ignores it. And it has to have a static class method New() and a constructor that takes no arguments. And when you use SetParameterCursorMovingCommand you pass in a pointer to an instance of your class, and the method name, which in this case should be "Update". You can also add more arguments to this method string, so long as your method also takes those additional arguments. And again, VTK wrapping is simple-minded, so the parameters should be of an C++ built-in scalar type or char *. Hope this helps... On 5/1/09 8:29 PM, "Nicolas Rannou" wrote: > Hi all, > > I'm currently developping an application for Slicer but errors occurs, > as if I didn't wrapped my application in TCL. > > I have an issue trying to use > vtkKWParameterValueFunctionEditor::SetParameterCursorMovingCommand(vtkObject > * object,const char * method). > I added a cursor in my histogram and I would like to call a function > each time I move the cursor. Which should be possible with the function > above. > So I created a function 'GetValue()' (to try). This function does not do > anything yet. > I called this function like this: > > /*******************************************************/ > ...-> SetParameterCursorMovingCommand( NULL, "GetValue"); > /*******************************************************/ > > The problem is that each time I move my cursor, an error occurs: > > /*******************************************************/ > ERROR: In > /projects/nacvisitors/nrannou/Slicer3/Slicer3-lib/KWWidgets/vtkKWTkUtilities.c > xx, > line 230 > vtkSlicerApplication (0xefa5b0): > Script: > GetValue > Returned Error on line 1: > invalid command name "GetValue" > Stack trace: > invalid command name "GetValue" > while executing > "GetValue" > > ERROR: In > /projects/nacvisitors/nrannou/Slicer3/Slicer3-lib/KWWidgets/vtkKWTkUtilities.c > xx, > line 230 > vtkSlicerApplication (0xefa5b0): > Script: > GetValue 960.666667 > Returned Error on line 1: > invalid command name "GetValue" > Stack trace: > invalid command name "GetValue" > while executing > "GetValue 960.666667" > /*******************************************************/ > > It seems like it is not wrapped. > The point is that I wrapped my application using in the CMakeList.txt: > > /*******************************************/ > include("${VTK_CMAKE_DIR}/vtkWrapTcl.cmake") > vtk_wrap_tcl3(EMSegment > EMSegment_TCL_SRCS > "${EMSegment_SRCS}" "") > /*******************************************/ > > And I couldn't find any documentation which helps about that. > > Thanks, > Nicolas > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject From pieper at bwh.harvard.edu Mon May 4 18:01:58 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 04 May 2009 18:01:58 -0400 Subject: [slicer-devel] Problen building Slicer3 for Visual Studio 2008 In-Reply-To: <368a82280905040422h9e15ffdm6a69045e26e1e0f8@mail.gmail.com> References: <368a82280905040422h9e15ffdm6a69045e26e1e0f8@mail.gmail.com> Message-ID: <49FF65D6.3050704@bwh.harvard.edu> Hi Ralph - Yes - I discovered this the other day but haven't had a chance to figure out a general purpose fix yet. Unlike the standard versions, it turns out that the free versions of the microsoft compilers won't upgrade from previous versions (python only ships with build files for Visual Studio 2003 .NET). I'm thinking I may need to put a set of upgraded project files for each version of VCExpress (2005 and 2008) into a patch that gets downloaded at build time (or something...any ideas?). In the near term, you could turn off python in slicer_variables.tcl or if you need python, follow the instructions for USE_SYSTEM_PYTHON described here: http://www.slicer.org/slicerWiki/index.php/Slicer3:Python Best, Steve Ralph Brecheisen wrote: > Hi, > > I'm following the build instructions for building Slicer3 on Windows > Vista and Visual Studio 2008 Express Edition. However, the > "getbuildtest.tcl" script I run from Cygwin at some point tries to > execute the following line: > > "running : {c:/Program Files/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpress.exe} PCBuild/pcbuild.sln /Upgrade" > > Visual Studio reports that it doesn't know the switch '/Upgrade' en > exits. The remainder of "getbuildtest.tcl" then also fails. > > Does anyone have experience with this problem as well? > > Thanks, > > Ralph > > > > > -- > Ir. Ralph Brecheisen > > Biomedical Image Analysis @ Technical University of Eindhoven, Den > Dolech 2, 5600 MB, Eindhoven, Netherlands > Image-Guided Neurosurgery @ Maastricht University Hospital > P. Debyelaan 25, 6229 HX, Maastricht, Netherlands > > Tel: +31-40-2475571 (Eindhoven) > Tel: +31-43-3876292 (Maastricht) > E-mail: r.brecheisen at tue.nl , > ralph.brecheisen at gmail.com > -------------------------------------------------------------------------------------------------- > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From invite+ihdf1i1e at facebookmail.com Tue May 5 08:28:23 2009 From: invite+ihdf1i1e at facebookmail.com (Ralph Brecheisen) Date: Tue, 5 May 2009 05:28:23 -0700 Subject: [slicer-devel] Check out my photos on Facebook Message-ID: <1be046920ede22a6ac61f89a8a3d4c3b@localhost.localdomain> Hi Slicer-devel, I set up a Facebook profile where I can post my pictures, videos and events and I want to add you as a friend so you can see it. First, you need to join Facebook! Once you join, you can also create your own profile. Thanks, Ralph To sign up for Facebook, follow the link below: http://www.facebook.com/p.php?i=557606169&k=32ET32S3V65M5EBDUC45VW&r slicer-devel at bwh.harvard.edu was invited to join Facebook by Ralph Brecheisen. If you do not wish to receive this type of email from Facebook in the future, please click on the link below to unsubscribe. http://www.facebook.com/o.php?k=5f8494&u=1764412273&mid=6a2e15G692acb71G0G8 Facebook's offices are located at 156 University Ave., Palo Alto, CA 94301. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090505/7a25c694/attachment.html From harish.slicer at gmail.com Wed May 6 13:15:39 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 6 May 2009 10:15:39 -0700 Subject: [slicer-devel] Register Images Module Message-ID: Hi all, I am trying to use the landmarks feature present in Register Images module. In the tab Advanced Initial Registration Parameters, can I give the fixed landmarks as vector of RAS values for eg: Fixed Landmarks : [0.8,12.1,25.0],[0.9,13.5,24.0] and moving landmarks as [0.8,28.1,30.1],[0.9,29.1,29.0]. Is the input format fine? Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090506/4d39faa9/attachment.html From harish.slicer at gmail.com Wed May 6 14:59:06 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Wed, 6 May 2009 11:59:06 -0700 Subject: [slicer-devel] automatic transform between two image volumes Message-ID: Hi all, Is there any module in slicer which can generate coordinate transforms between two image volumes. I mean given an input volume which needs to be tranformed to another volume, I want to get a tranformed file of this. Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090506/e78d7a96/attachment.html From pieper at bwh.harvard.edu Wed May 6 17:54:43 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Wed, 06 May 2009 17:54:43 -0400 Subject: [slicer-devel] [slicer-users] automatic transform between two image volumes In-Reply-To: References: Message-ID: <4A020723.9070809@bwh.harvard.edu> Hi Harish - You should be able to use the Filtering->Resample Scalar/Vector/DWI Volume module for that. -Steve Harish Doddi wrote: > > > Hi all, > > Is there any module in slicer which can generate coordinate transforms > between two image volumes. I mean given an input volume which needs to > be tranformed to another volume, I want to get a tranformed file of this. > > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From stephen.aylward at kitware.com Wed May 6 18:49:49 2009 From: stephen.aylward at kitware.com (Stephen Aylward) Date: Wed, 6 May 2009 18:49:49 -0400 Subject: [slicer-devel] [slicer-users] automatic transform between two image volumes In-Reply-To: <4A020723.9070809@bwh.harvard.edu> References: <4A020723.9070809@bwh.harvard.edu> Message-ID: <68a07c2d0905061549h355ec2e6iaec81c2d6459d789@mail.gmail.com> RegisterImages will also read and write transform files. Choose the transform file as the initial transform and set registration type to none. s On Wed, May 6, 2009 at 5:54 PM, Steve Pieper wrote: > Hi Harish - > > You should be able to use the Filtering->Resample Scalar/Vector/DWI > Volume module for that. > > -Steve > > Harish Doddi wrote: >> >> >> Hi all, >> >> Is there any module in slicer which can generate coordinate transforms >> between two image volumes. I mean given an input volume which needs to >> be tranformed to another volume, I want to get a tranformed file of this. >> >> Harish >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-users mailing list >> slicer-users at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >> To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > -- Stephen R. Aylward, Ph.D. Director of Medical Imaging Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 From ralph.brecheisen at gmail.com Fri May 8 10:00:16 2009 From: ralph.brecheisen at gmail.com (Ralph Brecheisen) Date: Fri, 8 May 2009 16:00:16 +0200 Subject: [slicer-devel] Problems running getbuildtest.tcl on Windows 7 Message-ID: <368a82280905080700kc3440b8sba7abd7033094476@mail.gmail.com> Hi, I know it might be a bit ambitious to try and install Slicer3 on Windows 7 RC but since I have had no problems so far with other software builds, I'd like to post a question anyway :) I followed the build instructions on this page http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions_Windows. I installed CMake 2.7 into the CMake-build directory. Also I installed Cygwin with the additional packages (svn, cvs, curl, tcltk, etc). However, upon running the getbuildtest.tcl script, the first thing that is complained about is the fact that the version of svn.exe is too old and should be updated. The remainder of the error messages are highly cryptic. I cannot find a single cause of the problem. On Windows Vista everything installed fine (except for having to disable Python in the slicer_variables.tcl). Can anyone help me with these issues? I attached the output of the build to this mail. Thanks in advance! Ralph -- "I respectfully decline the invitation to join your hallucination." - Scott Adams -- Ir. Ralph Brecheisen Biomedical Image Analysis @ Technical University of Eindhoven, Den Dolech 2, 5600 MB, Eindhoven, Netherlands Image-Guided Neurosurgery @ Maastricht University Hospital P. Debyelaan 25, 6229 HX, Maastricht, Netherlands Tel: +31-40-2475571 (Eindhoven) Tel: +31-43-3876292 (Maastricht) E-mail: r.brecheisen at tue.nl, ralph.brecheisen at gmail.com -------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090508/79ab3204/attachment.html -------------- next part -------------- More info: http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions_Windows - Install Cygwin with following packages - tcl/tk, svn, cvs, unzip, curl - Download Slicer3 sources - Disable Python in slicer_variables.tcl - Run 'getbuildtest.tcl' - This will fail From hayes at bwh.harvard.edu Fri May 8 13:24:33 2009 From: hayes at bwh.harvard.edu (Kathryn Hayes) Date: Fri, 8 May 2009 13:24:33 -0400 (EDT) Subject: [slicer-devel] Problems running getbuildtest.tcl on Windows 7 In-Reply-To: <368a82280905080700kc3440b8sba7abd7033094476@mail.gmail.com> References: <368a82280905080700kc3440b8sba7abd7033094476@mail.gmail.com> Message-ID: Hi Ralph, Which version of svn are you running? You can check by typing: svn --version at the Cygwin prompt. You might need to go into Cygwin Setup and get a newer svn. We're currently using 1.4.5. Thanks, Katie On Fri, 8 May 2009, Ralph Brecheisen wrote: > Hi, > > I know it might be a bit ambitious to try and install Slicer3 on Windows 7 > RC but since I have had no problems so far with other software builds, I'd > like to post a question anyway :) > I followed the build instructions on this page > http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions_Windows. > I installed CMake 2.7 into the CMake-build directory. > Also I installed Cygwin with the additional packages (svn, cvs, curl, tcltk, > etc). However, upon running the getbuildtest.tcl script, the first thing > that is complained about is the fact > that the version of svn.exe is too old and should be updated. The remainder > of the error messages are highly cryptic. I cannot find a single cause of > the problem. On Windows Vista > everything installed fine (except for having to disable Python in the > slicer_variables.tcl). > > Can anyone help me with these issues? I attached the output of the build to > this mail. > > Thanks in advance! > > Ralph > > > > From nrannou at bwh.harvard.edu Fri May 8 14:23:01 2009 From: nrannou at bwh.harvard.edu (Nicolas Rannou) Date: Fri, 08 May 2009 14:23:01 -0400 Subject: [slicer-devel] [slice-devel] Get the coordinate of a slice, in the red slice layout Message-ID: <4A047885.9020901@bwh.harvard.edu> Hi all, for my application, I load a volume, then, I would like to get the coordinates of the slice I can see in the red slice layout, to extract this slice from my volume. I assume it must be something like that: /---------------------------------------------------------------------/ ->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red")->...? /---------------------------------------------------------------------/ but I can't find the exact way to get it. Does somebody know how I can access to it? Thanks, Nicolas From harish.slicer at gmail.com Fri May 8 14:26:30 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Fri, 8 May 2009 11:26:30 -0700 Subject: [slicer-devel] Register Images Batch Make Message-ID: Hi all, I am trying to use batch processing for register images module. In the module Register Images Batch Make, I understand I need to give the directory for fixed and moving images. But I dont understand what is the moving image mask? Does the resample directory contain all output images of this module ? What does "batchmake" tab information indicate and what we should give as input? I am trying to do parameter exploration by using this module, but I am not sure how I can change the parameters. So if there is any documentation on how to use batch processing in slicer, kindly mail me. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090508/37efa764/attachment.html From fedorov at bwh.harvard.edu Fri May 8 16:36:28 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Fri, 8 May 2009 16:36:28 -0400 Subject: [slicer-devel] [slice-devel] Get the coordinate of a slice, in the red slice layout In-Reply-To: <4A047885.9020901@bwh.harvard.edu> References: <4A047885.9020901@bwh.harvard.edu> Message-ID: <82f536810905081336s10e1fecetbb2643dd583741a8@mail.gmail.com> Nicolas, You can use this to transform from XY slice coordinates to the RAS space: vtkSlicerSliceGUI *sliceGUI = vtkSlicerApplicationGUI::SafeDownCast( this->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red"); vtkMatrix4x4 *matrix = sliceGUI->GetLogic()->GetSliceNode()->GetXYToRAS(); hth AF On Fri, May 8, 2009 at 2:23 PM, Nicolas Rannou wrote: > Hi all, > > for my application, I load a volume, then, I would like to get the > coordinates of the slice I can see in the red slice layout, to extract > this slice from my volume. > I assume it must be something like that: > > /---------------------------------------------------------------------/ > ->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red")->...? > /---------------------------------------------------------------------/ > > but I can't find the exact way to get it. > Does somebody know how I can access to it? > > Thanks, > Nicolas > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From julien.finet at kitware.com Fri May 8 17:06:37 2009 From: julien.finet at kitware.com (Julien Finet) Date: Fri, 8 May 2009 17:06:37 -0400 Subject: [slicer-devel] Register Images Batch Make In-Reply-To: References: Message-ID: <5b9fa8230905081406j78509d00p16eea72defe2172e@mail.gmail.com> Hi Harish, Please find below the Slicer3.4 documention of the Register Image BatchMake module: http://www.slicer.org/slicerWiki/index.php/Modules:RegisterImagesBatch-Documentation-3.4 However I'm also currently writing some more detailed documentation. You can find the work in progress here: http://www.batchmake.org/Wiki/Batchmake_slicer To answer your question more specifically, the "Moving Image Mask" is a wildcard string that is used to select the "Moving Images" in the "Moving Image Directory" based on their name. The file name that match the mask will be processed as a moving image. Typically, your moving image mask should be *.mha or *.nrrd. The "Resample Image Mask" is a wildcard string that is used to generate the output file names. It can use the values of the XML variable names, typically it would be: "resampled-${movingImage}" I hope this helps. Don't hesitate to ask me more questions about it. I'll be happy to add more details in the documentation. Regards, Julien. On Fri, May 8, 2009 at 2:26 PM, Harish Doddi wrote: > > > Hi all, > > I am trying to use batch processing for register images module. In the > module Register Images Batch Make, > I understand I need to give the directory for fixed and moving images. But > I dont understand what is the moving image mask? Does the resample directory > contain all output images of this module ? What does "batchmake" tab > information indicate and what we should give as input? > > I am trying to do parameter exploration by using this module, but I am not > sure how I can change the parameters. So if there is any documentation on > how to use batch processing in slicer, kindly mail me. > > > Thanks, > Harish > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090508/cf3ba604/attachment.html From norman-k-williams at uiowa.edu Fri May 8 17:46:53 2009 From: norman-k-williams at uiowa.edu (kent williams) Date: Fri, 08 May 2009 16:46:53 -0500 Subject: [slicer-devel] How are fiducial points displayed manipulated Message-ID: Hey. For our non-slicer tracing program, I've been banging my head on using the vtkSeedWidget in order to allow points to be chosen for the purposes of registration, and other diverse purposes. Hans Johnson (my boss/colleague) said "can you manipulate the point in all 3 views?" (i.e. axial sagittal, coronal), and I had to say 'No, the Seed Widget is something associated with a single renderer. The points displayed on the off-plane views are actors without any corresponding interactor.' Hans Johnson just said 'well Slicer has fiducial points that you can move in any of the 3 planer views -- you should look at how Slicer does it.' So I just spent a half hour wandering around in the Slicer source without really seeing how Slicer does it. It obviously does not use the vtkSeedWidget! Anyone have a clue as to where to look to find the slicer Fiducial point stuff? From harish.slicer at gmail.com Fri May 8 18:19:51 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Fri, 8 May 2009 15:19:51 -0700 Subject: [slicer-devel] Register Images Batch Make In-Reply-To: <5b9fa8230905081406j78509d00p16eea72defe2172e@mail.gmail.com> References: <5b9fa8230905081406j78509d00p16eea72defe2172e@mail.gmail.com> Message-ID: Hi Julien, Thanks for the information. It is very helpful. I want to ask one more question. I see from the information given that you can register several moving images to one fixed image. But my aim is that I have one fixed image, one moving image and want to experiment various possible parameters and get various results (the whole process is to find which parameters give the best registration). So is there any module in slicer where we can do something like this or should I do it through commmand line interface ? Thanks, Harish On Fri, May 8, 2009 at 2:06 PM, Julien Finet wrote: > Hi Harish, > > Please find below the Slicer3.4 documention of the Register Image BatchMake > module: > > http://www.slicer.org/slicerWiki/index.php/Modules:RegisterImagesBatch-Documentation-3.4 > > However I'm also currently writing some more detailed documentation. You > can find the work in progress here: > http://www.batchmake.org/Wiki/Batchmake_slicer > > To answer your question more specifically, the "Moving Image Mask" is a > wildcard string that is used to select the "Moving Images" in the "Moving > Image Directory" based on their name. The file name that match the mask will > be processed as a moving image. Typically, your moving image mask should be > *.mha or *.nrrd. > The "Resample Image Mask" is a wildcard string that is used to generate the > output file names. It can use the values of the XML variable names, > typically it would be: "resampled-${movingImage}" > > I hope this helps. Don't hesitate to ask me more questions about it. I'll > be happy to add more details in the documentation. > > Regards, > Julien. > > On Fri, May 8, 2009 at 2:26 PM, Harish Doddi wrote: > >> >> >> Hi all, >> >> I am trying to use batch processing for register images module. In the >> module Register Images Batch Make, >> I understand I need to give the directory for fixed and moving images. But >> I dont understand what is the moving image mask? Does the resample directory >> contain all output images of this module ? What does "batchmake" tab >> information indicate and what we should give as input? >> >> I am trying to do parameter exploration by using this module, but I am not >> sure how I can change the parameters. So if there is any documentation on >> how to use batch processing in slicer, kindly mail me. >> >> >> Thanks, >> Harish >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to >> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >> subject >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090508/3da6951c/attachment.html From julien.finet at kitware.com Sat May 9 09:31:39 2009 From: julien.finet at kitware.com (Julien Finet) Date: Sat, 9 May 2009 09:31:39 -0400 Subject: [slicer-devel] Register Images Batch Make Message-ID: <1FA3D11A-58E8-4E4E-B68A-440C1F7EE2DA@kitware.com> Harish, You can definitely do that with Register Images BatchMake module. Right now the module XML has been written to vary on the inputs and to vary on the number of iterations of the rigid, affine or bspline method. If you set a min, max and step, the module will automatically create a for-loop and run the register images module max-min / step times with different number of iterations value each time. Feel free to modify the Register Images BatchMake module XML file to vary other parameters.You just have to follow the min/max/step mechanism. Take example on the rigid, affine or bspline number of iterations parameters. Regards, Julien. On May 8, 2009, at 6:19 PM, Harish Doddi wrote: > Hi Julien, > > Thanks for the information. It is very helpful. I want to ask one > more question. I see from the information given that you can > register several moving images to one fixed image. But my aim is > that I have one fixed image, one moving image and want to experiment > various possible parameters and get various results (the whole > process is to find which parameters give the best registration). So > is there any module in slicer where we can do something like this or > should I do it through commmand line interface ? > > > Thanks, > Harish > > On Fri, May 8, 2009 at 2:06 PM, Julien Finet > wrote: > Hi Harish, > > Please find below the Slicer3.4 documention of the Register Image > BatchMake module: > http://www.slicer.org/slicerWiki/index.php/Modules:RegisterImagesBatch-Documentation-3.4 > > However I'm also currently writing some more detailed documentation. > You can find the work in progress here: > http://www.batchmake.org/Wiki/Batchmake_slicer > > To answer your question more specifically, the "Moving Image Mask" > is a wildcard string that is used to select the "Moving Images" in > the "Moving Image Directory" based on their name. The file name that > match the mask will be processed as a moving image. Typically, your > moving image mask should be *.mha or *.nrrd. > The "Resample Image Mask" is a wildcard string that is used to > generate the output file names. It can use the values of the XML > variable names, typically it would be: "resampled-${movingImage}" > > I hope this helps. Don't hesitate to ask me more questions about it. > I'll be happy to add more details in the documentation. > > Regards, > Julien. > > On Fri, May 8, 2009 at 2:26 PM, Harish Doddi > wrote: > > > Hi all, > > I am trying to use batch processing for register images module. In > the module Register Images Batch Make, > I understand I need to give the directory for fixed and moving > images. But I dont understand what is the moving image mask? Does > the resample directory contain all output images of this module ? > What does "batchmake" tab information indicate and what we should > give as input? > > I am trying to do parameter exploration by using this module, but I > am not sure how I can change the parameters. So if there is any > documentation on how to use batch processing in slicer, kindly mail > me. > > > Thanks, > Harish > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090509/0c86c8eb/attachment-0001.html From pieper at bwh.harvard.edu Mon May 11 09:36:21 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 11 May 2009 09:36:21 -0400 Subject: [slicer-devel] How are fiducial points displayed manipulated In-Reply-To: References: Message-ID: <4A0829D5.3080706@bwh.harvard.edu> Hi Kent - The short answer is you can have a look at Base/GUI/Tcl/SeedSWidget.tcl and FiducialsSWidget.tcl. Longer answer is that for the 2D slice viewers in slicer, the customization of VTK for slicer's purposes is implemented in incrTcl (object oriented tcl) with a base class called SWidget (Slicer Widget). These classes make the link between vtk-based primitives (events, actors, etc) and slicer's application structures like vtkSlicerSliceGUI, vtkMRMLFiducialList, etc. The main reasons for this were developer preference (mine!) for being able to reload the tcl scripts while the program is running in order to redefine the behavior of the widgets and because the vtk widgets didn't have the basic behavior I needed (like they didn't highlight on mouseover, which always bothered me). SWidget subclasses are also heavily used in the Editor module. In the 3D viewer, we do use vtk widgets, for example, Base/GUI/vtkSlicerFidudicalListWidget uses vtkPointWidgets for 3D interaction. Also, Base/GUI/Tcl/SlicePlaneSWidget.tcl manages a 3D vtk widget for interactive slice plane manipulation. In the end, since slicer internally considers the MRML representation as the 'definitive owner' of the fiducials state, there can be many different graphical editors (SWidgets in 2D, PointWidgets in 3D, and KWWidgets in the Fiducials Module). As long as they all track events correctly they cooperate. HTH, -Steve kent williams wrote: > Hey. For our non-slicer tracing program, I've been banging my head on using > the vtkSeedWidget in order to allow points to be chosen for the purposes of > registration, and other diverse purposes. > > Hans Johnson (my boss/colleague) said "can you manipulate the point in all 3 > views?" (i.e. axial sagittal, coronal), and I had to say 'No, the Seed > Widget is something associated with a single renderer. The points displayed > on the off-plane views are actors without any corresponding interactor.' > > Hans Johnson just said 'well Slicer has fiducial points that you can move in > any of the 3 planer views -- you should look at how Slicer does it.' > > So I just spent a half hour wandering around in the Slicer source without > really seeing how Slicer does it. It obviously does not use the > vtkSeedWidget! > > Anyone have a clue as to where to look to find the slicer Fiducial point > stuff? > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From karthik.krishnan at kitware.com Mon May 11 09:51:13 2009 From: karthik.krishnan at kitware.com (Karthik Krishnan) Date: Mon, 11 May 2009 09:51:13 -0400 Subject: [slicer-devel] How are fiducial points displayed manipulated In-Reply-To: <4A0829D5.3080706@bwh.harvard.edu> References: <4A0829D5.3080706@bwh.harvard.edu> Message-ID: <9ddb27260905110651i1e3817e2he1916c3d8142a719@mail.gmail.com> Kent: To answer your question specifically, as you've pointed out, VTK widgets can only be associated with a single renderwindowinteractor, which in turn is associated with a single render window. Slicer essentially uses a Model View design pattern. The MRML representation is the model and the views (render windows) listen to events etc to coordinate. In particular the hierarchy looks like vtkMRMLScene | ---- vtkMRMLFiducialListNode | ------ fiducials As Steve already pointed out, the 2D fiducials do not use the VTK widgets, although I don't think there are significant deficiencies in the VTK widgets. Highlighting as well as cursor change of nodes on mouse over can be quite easily built into the VTK widgets. Several widgets do this, for instance, vtkParallelopipedWidget, vtkContourWidget etc. The 3D ones use the old vtkPointWidget. In the end, you will probably need to write your own model, much like MRML, to listen to the VTK widgets for interaction events and update all the other widgets that represent the same model in different views (render windows) yourself. You may reuse the MRML library, if its generic enough to suite your needs. On Mon, May 11, 2009 at 9:36 AM, Steve Pieper wrote: > Hi Kent - > > The short answer is you can have a look at Base/GUI/Tcl/SeedSWidget.tcl > and FiducialsSWidget.tcl. > > Longer answer is that for the 2D slice viewers in slicer, the > customization of VTK for slicer's purposes is implemented in incrTcl > (object oriented tcl) with a base class called SWidget (Slicer Widget). > These classes make the link between vtk-based primitives (events, > actors, etc) and slicer's application structures like vtkSlicerSliceGUI, > vtkMRMLFiducialList, etc. The main reasons for this were developer > preference (mine!) for being able to reload the tcl scripts while the > program is running in order to redefine the behavior of the widgets and > because the vtk widgets didn't have the basic behavior I needed (like > they didn't highlight on mouseover, which always bothered me). SWidget > subclasses are also heavily used in the Editor module. > > In the 3D viewer, we do use vtk widgets, for example, > Base/GUI/vtkSlicerFidudicalListWidget uses vtkPointWidgets for 3D > interaction. Also, Base/GUI/Tcl/SlicePlaneSWidget.tcl manages a 3D vtk > widget for interactive slice plane manipulation. > > In the end, since slicer internally considers the MRML representation as > the 'definitive owner' of the fiducials state, there can be many > different graphical editors (SWidgets in 2D, PointWidgets in 3D, and > KWWidgets in the Fiducials Module). As long as they all track events > correctly they cooperate. > > HTH, > -Steve > > > > kent williams wrote: > > Hey. For our non-slicer tracing program, I've been banging my head on > using > > the vtkSeedWidget in order to allow points to be chosen for the purposes > of > > registration, and other diverse purposes. > > > > Hans Johnson (my boss/colleague) said "can you manipulate the point in > all 3 > > views?" (i.e. axial sagittal, coronal), and I had to say 'No, the Seed > > Widget is something associated with a single renderer. The points > displayed > > on the off-plane views are actors without any corresponding interactor.' > > > > Hans Johnson just said 'well Slicer has fiducial points that you can move > in > > any of the 3 planer views -- you should look at how Slicer does it.' > > > > So I just spent a half hour wandering around in the Slicer source without > > really seeing how Slicer does it. It obviously does not use the > > vtkSeedWidget! > > > > Anyone have a clue as to where to look to find the slicer Fiducial point > > stuff? > > > > _______________________________________________ > > slicer-devel mailing list > > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > -- Karthik Krishnan R&D Engineer, Kitware Inc. Ph: 518 881 4919 Fax: 518 371 4573 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/c69a7427/attachment.html From aescoto2 at uwo.ca Mon May 11 11:35:21 2009 From: aescoto2 at uwo.ca (Abelardo Escoto) Date: Mon, 11 May 2009 11:35:21 -0400 Subject: [slicer-devel] Aurora tracker module help Message-ID: Hello: This is Abelardo Escoto, I am a Research Assistant working at CSTAR. Currently we are trying to develop some application using slicer3D, reason what I have been learning about it. I have some questions related with the IGT:Navigation-with-Aurora (http://wiki.na-mic.org/Wiki/index.php/IGT:ToolKit/Navigation-with-Aurora). 1- In Step 3 it says that we could use the binaries, however, there is no binary for windows, OS that I am using. Is there any binary available for windows? 2- I am trying to build IGSTK. I think that I could do it properly, I followed all the steps but, when building the IGSTK with VS it does not compiled. However it did compile when building IGSTKSandbox. Is this normal/ok? 3- Once the IGSTKSandbox has been compiled with VS, what am I suppose to do in order to use IGSTK? 4- I am following the tutorial Advanced Tutorial Image Guided Therapy in Slicer3 (http://wiki.na-mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf). All the information until slide 54 are pretty clear, however on slide 55 it says: "run the IGSTK test program...". My question is: where is that program?? I tried to use the following path that is shown on that slide: $ ./bin/igstkSandboxTests igstkAuroraTrackerToolObserverToOpenIGTLinkRelayTest localhost 18944 10000 0 1 but is does not work. 5- Is there any other module/program available to connect the Aurora tracker with slicer3? Thank you very much for your time and attention, Abelardo Escoto -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/8c6f814b/attachment.html From dpace at imaging.robarts.ca Mon May 11 11:50:06 2009 From: dpace at imaging.robarts.ca (Danielle Pace) Date: Mon, 11 May 2009 11:50:06 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: Message-ID: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> Hi Abelardo, 1) I don't think we have a Windows binary for IGSTK Sandbox, so you would have to build IGSTK and IGSTK Sandbox from source. 2) If IGSTK didn't build, then you should resolve it - can you post the error messages? 3) For interfacing with the Aurora tracking system, once you have IGSTK and IGSTKSandbox built correctly then you just need to run igstkAuroraTrackerToolObserverToOpenIGTLinkRelayTest to interface the Aurora with 3D Slicer - this test program is the code that uses IGSTK to enable tracking within Slicer. 4) If you navigate to the IGSTKSandbox-build directory, where you built IGSTKSandbox, the IGSTK test program will be located in bin/igstkSandboxTests. 5) I'm not aware of any, but perhaps Junichi or Haiying would be able to comment on if there have been any updates since the advanced navigation tutorial was made? Hope that helps, Danielle -- Danielle Pace, BCmpH Masters Candidate, Biomedical Engineering Imaging Research Laboratories, Robarts Research Institute The University of Western Ontario, London, Ontario, Canada _____ From: slicer-devel-bounces at bwh.harvard.edu [mailto:slicer-devel-bounces at bwh.harvard.edu] On Behalf Of Abelardo Escoto Sent: May 11, 2009 11:35 AM To: Developer, Slicer Subject: [slicer-devel] Aurora tracker module help Hello: This is Abelardo Escoto, I am a Research Assistant working at CSTAR. Currently we are trying to develop some application using slicer3D, reason what I have been learning about it. I have some questions related with the IGT:Navigation-with-Aurora (http://wiki.na-mic.org/Wiki/index.php/IGT:ToolKit/Navigation-with-Aurora). 1- In Step 3 it says that we could use the binaries, however, there is no binary for windows, OS that I am using. Is there any binary available for windows? 2- I am trying to build IGSTK. I think that I could do it properly, I followed all the steps but, when building the IGSTK with VS it does not compiled. However it did compile when building IGSTKSandbox. Is this normal/ok? 3- Once the IGSTKSandbox has been compiled with VS, what am I suppose to do in order to use IGSTK? 4- I am following the tutorial Advanced Tutorial Image Guided Therapy in Slicer3 (http://wiki.na-mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf). All the information until slide 54 are pretty clear, however on slide 55 it says: "run the IGSTK test program...". My question is: where is that program?? I tried to use the following path that is shown on that slide: $ ./bin/igstkSandboxTests igstkAuroraTrackerToolObserverToOpenIGTLinkRelayTest localhost 18944 10000 0 1 but is does not work. 5- Is there any other module/program available to connect the Aurora tracker with slicer3? Thank you very much for your time and attention, Abelardo Escoto -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/f5ab1cb7/attachment-0001.html From aescoto2 at uwo.ca Mon May 11 12:10:29 2009 From: aescoto2 at uwo.ca (Abelardo Escoto) Date: Mon, 11 May 2009 12:10:29 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> Message-ID: An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/980a8dc7/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: IGSTK build errors.JPG Type: image/jpeg Size: 150714 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/980a8dc7/attachment-0001.jpe From harish.slicer at gmail.com Mon May 11 13:55:20 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 11 May 2009 10:55:20 -0700 Subject: [slicer-devel] Register Images batch make module Message-ID: Hi all, I am trying to use the registerd batch make module to find the optimum parameters for registering a moving image to a fixed image. I have my image datasets in .mha format. I have give the path for the fixed image file and moving image directory and I gave moving image mask as *.mha. For Resample Image, I created a directory and gave that directory as input. For Resample Image mask I gave I give the input as Registered-${affineMaxIterations} since I want to find the resulting image for each possible no of iterations ranging from affineIterations_min to affineIterations_max in steps of affineIterations_step. For eg if I give min value as 50 and max value 100 in steps of 10, I want the resulting image for 50, 60, 70, 80, 90, 100.. But I dont understand why I get the following error in the output. Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Error: Could not run RegisterImagesPath-NOTFOUND: No such file or directory Kindly guide me if I am missing something. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/1d7f2ac0/attachment.html From harish.slicer at gmail.com Mon May 11 13:58:33 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 11 May 2009 10:58:33 -0700 Subject: [slicer-devel] Register image parameters Message-ID: Hi all, I understand from the documentation that the parameters rigid sampling ratio, affine sampling ratio etc denote "the Portion of the image pixels to be used when computing the metric" So does this value range from 0-1. I am not getting good results for my dataset with default values. So can I give some higher values like 0.6, 0.8 etc or are there any side effects because of this. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/230f1b36/attachment.html From raquel.torresperalta at gmail.com Mon May 11 14:41:07 2009 From: raquel.torresperalta at gmail.com (Raquel Torres Peralta) Date: Mon, 11 May 2009 11:41:07 -0700 Subject: [slicer-devel] Compiling Slicer In-Reply-To: <81869fe90905111136y6105b59di1dbc6edd7752d3fc@mail.gmail.com> References: <81869fe90905111136y6105b59di1dbc6edd7752d3fc@mail.gmail.com> Message-ID: <81869fe90905111141v6a5a45b5v93e936677624fc6d@mail.gmail.com> Hello everyone. I?m trying to compile the 3D Slicer. I?ve solved some errors but now the process stops with no more info of what?s happening so I don?t know what to do. running: {C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpre ss.exe} Slicer3.SLN /build Debug child process exited abnormally Here?s what I have done so far for solving previous errors: - Run the compilation process using the administrator account and disable the ?Turn User Account Control on or off". - I?ve added the next line to the slicer_variables.tcl file in Slicer3 and also in slicer_variables.tcl file in Slicer3_2 to specify the location of the building tool: if { [file exists "C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpress.exe"] } { set ::GENERATOR "Visual Studio 9 2008" set ::MAKE "C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpress.exe" set ::COMPILER_PATH "C:/Program Files (x86)/Microsoft Visual Studio 9.0/VC/bin" Next I got the error: CMake Error: The source "D:/cygwin/home/Raquel/Slicer3/CMakeLists.txt" does not match the source "D:/cygwin/home/Raquel/Slicer-3-2/CMakeLists.txt" used to generate cache. Re-run cmake with a different source directory. running: {C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpress.exe} Slicer3.SLN /build Debug child process exited abnormally Then I erased the CMakeCache.txt in my building directory and tried again. Is then when I got the error: running: {C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpre ss.exe} Slicer3.SLN /build Debug ?Here it stops for a while - child process exited abnormally My OS: Windows Vista Compiler: Microsoft Visual C++ 2008 Express Edition CMake 2.7 Do you know what?s happening? Best regards, Raquel Here?s a copy of the messages: $ ./Slicer3/Scripts/getbuildtest.tcl -t "" getbuildtest: setting build list to Sourcing D:/cygwin/home/Raquel/Slicer3/slicer_variables.tcl Slicer3_HOME is D:/cygwin/home/Raquel/Slicer3 making with C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExp ress.exe running: svn switch http://svn.slicer.org/Slicer3/branches/Slicer-3-2 At revision 9444. running: sh ./Scripts/genlib.tcl D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib Stripped args = D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib Sourcing D:/cygwin/home/Raquel/Slicer3/slicer_variables.tcl running: svn co http://svn.slicer.org/Slicer3-lib-mirrors/trunk/Binaries/Windows /python-build python-build Checked out revision 128. running: cvs -d :pserver:anoncvs:@www.kwwidgets.org:/cvsroot/KWWidgets login Logging in to :pserver:anoncvs at www.kwwidgets.org:2401/cvsroot/KWWidgets running: cvs -q -z3 -d :pserver:anoncvs at www.kwwidgets.org:/cvsroot/KWWidgets che ckout -r Slicer-3-2 KWWidgets running: D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe {-GVisual Studio 9 2008} -DVTK_DIR:PATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3 -lib/VTK-build {-DCMAKE_CXX_COMPILER:STRING=C:/Program Files (x86)/Microsoft Vis ual Studio 9.0/VC/bin/cl} {-DCMAKE_CXX_COMPILER_FULLPATH:FILEPATH=C:/Program Fil es (x86)/Microsoft Visual Studio 9.0/VC/bin/cl} -DBUILD_SHARED_LIBS:BOOL=ON -DCM AKE_SKIP_RPATH:BOOL=ON -DBUILD_EXAMPLES:BOOL=OFF -DKWWidgets_BUILD_EXAMPLES:BOOL =OFF -DBUILD_TESTING:BOOL=OFF -DKWWidgets_BUILD_TESTING:BOOL=OFF -DCMAKE_BUILD_T YPE:STRING=Debug ../KWWidgets -- Configuring done -- Generating done -- Build files have been written to: D:/cygwin/home/Raquel/Slicer3-lib/KWWidgets -build running: {C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpre ss.exe} KWWidgets.SLN /build Debug All packages compiled. Slicer3_HOME is D:/cygwin/home/Raquel/Slicer3 cvs checkout: closing down connection to www.kwwidgets.org child process exited abnormally running: D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe {-GVisual Studio 9 2008} {-DMAKECOMMAND:STRING=C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpress.exe} {-DCMAKE_CXX_COMPILER:STRING=C:/Pro gram Files (x86)/Microsoft Visual Studio 9.0/VC/bin/cl} {-DCMAKE_CXX_COMPILER_FU LLPATH:FILEPATH=C:/Program Files (x86)/Microsoft Visual Studio 9.0/VC/bin/cl} -D ITK_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/Insight-build -DKW Widgets_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/KWWidgets-buil d -DTEEM_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/teem-build -D IGSTK_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/IGSTK-build -DIN CR_TCL_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/tcl-build/l ib/lib/itcl3.2/itcl32.lib -DINCR_TK_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slice r3/../Slicer3-lib/tcl-build/lib/lib/itk3.2/itk32.lib -DSlicer3_USE_PYTHON=OFF -D PYTHON_INCLUDE_PATH:PATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/python-bui ld/include -DPYTHON_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-li b/python-build/Libs/python25.lib -DSandBox_DIR:FILEPATH=D:/cygwin/home/Raquel/Sl icer3/../Slicer3-lib/NAMICSandBox -DCMAKE_BUILD_TYPE=Debug -DSlicer3_VERSION_PAT CH:STRING=2009-05-10 -DCPACK_GENERATOR:STRING=NSIS -DCPACK_PACKAGE_FILE_NAME:STR ING=Slicer3-3.2.2009-05-10-win32 -DSlicer3_USE_IGSTK=OFF -DSlicer3_USE_NAVITRACK =OFF -DNAVITRACK_LIB_DIR:FILEPATH= -DNAVITRACK_INC_DIR:FILEPATH= -DSLICERLIBCURL _DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/cmcurl-build -DCMAKE_ VERBOSE_MAKEFILE:BOOL=OFF D:/cygwin/home/Raquel/Slicer3 -- Configuring done -- Generating done -- Build files have been written to: D:/cygwin/home/Raquel/Slicer3-build running: D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe {-DMAKECOMMAND:STRING=C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7 /IDE/VCExpress.exe Slicer3.sln /build Debug /project ALL_BUILD} D:/cygwin/home/R aquel/Slicer3 -- Configuring done -- Generating done -- Build files have been written to: D:/cygwin/home/Raquel/Slicer3-build running: {C:/Program Files (x86)/Microsoft Visual Studio 9.0/Common7/IDE/VCExpre ss.exe} Slicer3.SLN /build Debug child process exited abnormally -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/96b1fbff/attachment-0001.html From harish.slicer at gmail.com Mon May 11 14:43:36 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 11 May 2009 11:43:36 -0700 Subject: [slicer-devel] Reg Register Images Landmark Parameter Message-ID: Hi all, I am trying to use the register image module with landmarks as parameter. I see the tutorial says landmark as " A vector string (comma separated base-3 list) of the indexes of corresponding points in the fixed and moving images" >From what I understand, I am trying to plot fiducial points on the fixed image and moving image so that one point maps to other. Can I give these fiducial points as input to these landmarks ie if in fixed image I have one point as [-15.314, 2.0, 4.5] and moving image as [-10.0, 4, 3.5] Can I give both these points as input to fixed/moving landmarks respectively or is there something I am missing? Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/1782f9ed/attachment.html From harish.slicer at gmail.com Mon May 11 14:49:29 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Mon, 11 May 2009 11:49:29 -0700 Subject: [slicer-devel] Compiling Slicer In-Reply-To: <81869fe90905111141v6a5a45b5v93e936677624fc6d@mail.gmail.com> References: <81869fe90905111136y6105b59di1dbc6edd7752d3fc@mail.gmail.com> <81869fe90905111141v6a5a45b5v93e936677624fc6d@mail.gmail.com> Message-ID: Hi Raquel, I am not sure if this problem is the same I faced before. Anyhow, I saw these error messages arising because of the python build that is being built as part of slicer. There is a bug in slicer while integrating python modules. Open Slicer3/slicer_variables.tcl Search for string "USE_PYTHON" Set this string to OFF (Default is ON) Also set the string "USE_NUMPY" to OFF(Default is ON). and re-compile.. Hope this helps. Thanks, Harish On Mon, May 11, 2009 at 11:41 AM, Raquel Torres Peralta < raquel.torresperalta at gmail.com> wrote: > > Hello everyone. > > I?m trying to compile the 3D Slicer. I?ve solved some errors but now the > process stops with no more info of what?s happening so I don?t know what to > do. > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} Slicer3.SLN /build Debug > > > > child process exited abnormally > > > > Here?s what I have done so far for solving previous errors: > > - Run the compilation process using the administrator account and > disable the ?Turn User Account Control on or off". > > - I?ve added the next line to the slicer_variables.tcl file in > Slicer3 and also in slicer_variables.tcl file in Slicer3_2 to specify the > location of the building tool: > > > > if { [file exists "C:/Program Files (x86)/Microsoft Visual > Studio 9.0/Common7/IDE/VCExpress.exe"] } { set ::GENERATOR "Visual > Studio 9 2008" > > set ::MAKE "C:/Program Files (x86)/Microsoft Visual > Studio 9.0/Common7/IDE/VCExpress.exe" > > set ::COMPILER_PATH "C:/Program Files (x86)/Microsoft Visual > Studio 9.0/VC/bin" > > > > Next I got the error: > > CMake Error: The source "D:/cygwin/home/Raquel/Slicer3/CMakeLists.txt" does > not match the source "D:/cygwin/home/Raquel/Slicer-3-2/CMakeLists.txt" used > to generate cache. Re-run cmake with a different source directory. > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpress.exe} Slicer3.SLN /build Debug > > > > child process exited abnormally > > > > Then I erased the CMakeCache.txt in my building directory and tried again. > Is then when I got the error: > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} Slicer3.SLN /build Debug ?Here it stops for a while - > > > > child process exited abnormally > > > > > > My OS: Windows Vista > > Compiler: Microsoft Visual C++ 2008 Express Edition > > CMake 2.7 > > > > Do you know what?s happening? > > Best regards, > > Raquel > > > > Here?s a copy of the messages: > > > > $ ./Slicer3/Scripts/getbuildtest.tcl -t "" > > getbuildtest: setting build list to > > Sourcing D:/cygwin/home/Raquel/Slicer3/slicer_variables.tcl > > Slicer3_HOME is D:/cygwin/home/Raquel/Slicer3 > > making with C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExp > > ress.exe > > running: svn switch http://svn.slicer.org/Slicer3/branches/Slicer-3-2 > > At revision 9444. > > > > running: sh ./Scripts/genlib.tcl > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib > > Stripped args = D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib > > Sourcing D:/cygwin/home/Raquel/Slicer3/slicer_variables.tcl > > running: svn co > http://svn.slicer.org/Slicer3-lib-mirrors/trunk/Binaries/Windows > > /python-build python-build > > Checked out revision 128. > > > > running: cvs -d :pserver:anoncvs:@www.kwwidgets.org:/cvsroot/KWWidgets > login > > Logging in to :pserver:anoncvs at www.kwwidgets.org:2401/cvsroot/KWWidgets > > > > running: cvs -q -z3 -d :pserver:anoncvs at www.kwwidgets.org:/cvsroot/KWWidgets > che > > ckout -r Slicer-3-2 KWWidgets > > > > running: > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe > > {-GVisual Studio 9 2008} > -DVTK_DIR:PATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3 > > -lib/VTK-build {-DCMAKE_CXX_COMPILER:STRING=C:/Program Files > (x86)/Microsoft Vis > > ual Studio 9.0/VC/bin/cl} > {-DCMAKE_CXX_COMPILER_FULLPATH:FILEPATH=C:/Program Fil > > es (x86)/Microsoft Visual Studio 9.0/VC/bin/cl} -DBUILD_SHARED_LIBS:BOOL=ON > -DCM > > AKE_SKIP_RPATH:BOOL=ON -DBUILD_EXAMPLES:BOOL=OFF > -DKWWidgets_BUILD_EXAMPLES:BOOL > > =OFF -DBUILD_TESTING:BOOL=OFF -DKWWidgets_BUILD_TESTING:BOOL=OFF > -DCMAKE_BUILD_T > > YPE:STRING=Debug ../KWWidgets > > -- Configuring done > > -- Generating done > > -- Build files have been written to: > D:/cygwin/home/Raquel/Slicer3-lib/KWWidgets > > -build > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} KWWidgets.SLN /build Debug > > > > All packages compiled. > > > > Slicer3_HOME is D:/cygwin/home/Raquel/Slicer3 > > cvs checkout: closing down connection to www.kwwidgets.org > > child process exited abnormally > > running: > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe > > {-GVisual Studio 9 2008} {-DMAKECOMMAND:STRING=C:/Program Files > (x86)/Microsoft > > Visual Studio 9.0/Common7/IDE/VCExpress.exe} > {-DCMAKE_CXX_COMPILER:STRING=C:/Pro > > gram Files (x86)/Microsoft Visual Studio 9.0/VC/bin/cl} > {-DCMAKE_CXX_COMPILER_FU > > LLPATH:FILEPATH=C:/Program Files (x86)/Microsoft Visual Studio > 9.0/VC/bin/cl} -D > > ITK_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/Insight-build > -DKW > > > Widgets_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/KWWidgets-buil > > d > -DTEEM_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/teem-build > -D > > IGSTK_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/IGSTK-build > -DIN > > > CR_TCL_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/tcl-build/l > > ib/lib/itcl3.2/itcl32.lib > -DINCR_TK_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slice > > r3/../Slicer3-lib/tcl-build/lib/lib/itk3.2/itk32.lib > -DSlicer3_USE_PYTHON=OFF -D > > > PYTHON_INCLUDE_PATH:PATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/python-bui > > ld/include > -DPYTHON_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-li > > b/python-build/Libs/python25.lib > -DSandBox_DIR:FILEPATH=D:/cygwin/home/Raquel/Sl > > icer3/../Slicer3-lib/NAMICSandBox -DCMAKE_BUILD_TYPE=Debug > -DSlicer3_VERSION_PAT > > CH:STRING=2009-05-10 -DCPACK_GENERATOR:STRING=NSIS > -DCPACK_PACKAGE_FILE_NAME:STR > > ING=Slicer3-3.2.2009-05-10-win32 -DSlicer3_USE_IGSTK=OFF > -DSlicer3_USE_NAVITRACK > > =OFF -DNAVITRACK_LIB_DIR:FILEPATH= -DNAVITRACK_INC_DIR:FILEPATH= > -DSLICERLIBCURL > > _DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/cmcurl-build > -DCMAKE_ > > VERBOSE_MAKEFILE:BOOL=OFF D:/cygwin/home/Raquel/Slicer3 > > -- Configuring done > > -- Generating done > > -- Build files have been written to: D:/cygwin/home/Raquel/Slicer3-build > > > > running: > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe > > {-DMAKECOMMAND:STRING=C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7 > > /IDE/VCExpress.exe Slicer3.sln /build Debug /project ALL_BUILD} > D:/cygwin/home/R > > aquel/Slicer3 > > -- Configuring done > > -- Generating done > > -- Build files have been written to: D:/cygwin/home/Raquel/Slicer3-build > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} Slicer3.SLN /build Debug > > > > child process exited abnormally > > > > > > > > > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the > subject > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/30824cc6/attachment.html From stephen.aylward at kitware.com Mon May 11 15:12:13 2009 From: stephen.aylward at kitware.com (Stephen Aylward) Date: Mon, 11 May 2009 15:12:13 -0400 Subject: [slicer-devel] Register image parameters In-Reply-To: References: Message-ID: <68a07c2d0905111212k7a08c2d4m748dd75dd011d730@mail.gmail.com> You can increase the sampling...but it may not help. Really depends upon your data (e.g., sparseness of features wrt the metric being used) as well as the transform (e.g., dimensionality). As well as what you mean by "good results" :) The ITK Software Guide has a great chapter on image registration. Covers topics like transforms, optimizers, parameter selection, etc. You can get a free copy online at: http://www.itk.org/ItkSoftwareGuide.pdf As a follow-on book, I suggest Insight Into Images book: http://www.amazon.com/Insight-into-Images-Segmentation-Registration/dp/1568812175 Otherwise, perhaps post more info on the specific problems you are having and perhaps we can provide some suggestions. Stephen On Mon, May 11, 2009 at 1:58 PM, Harish Doddi wrote: > > > Hi all, > > I understand from the documentation that the parameters rigid sampling > ratio, affine sampling ratio etc denote "the Portion of the image pixels to > be used when computing the metric" > > So does this value range from 0-1. I am not getting good results for my > dataset with default values. So can I give some higher values like 0.6, 0.8 > etc or are there any side effects because of this. > > > Thanks, > Harish > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > -- Stephen R. Aylward, Ph.D. Director of Medical Imaging Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 From hliu at bwh.harvard.edu Mon May 11 15:37:34 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Mon, 11 May 2009 15:37:34 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: Message-ID: <4A087E7E.8010100@bwh.harvard.edu> Hi Abelardo, I embed my answers below. Abelardo Escoto wrote: > Hello: > > This is Abelardo Escoto, I am a Research Assistant working at CSTAR. > Currently we are trying to develop some application using slicer3D, > reason what I have been learning about it. > > I have some questions related with the IGT:Navigation-with-Aurora > (http://wiki.na-mic.org/Wiki/index.php/IGT:ToolKit/Navigation-with-Aurora). > 1- In Step 3 it says that we could use the binaries, however, there is > no binary for windows, OS that I am using. Is there any binary > available for windows? The tutorial was written a little while ago. On that wiki page, only Linux binary of IGSTK was uploaded. You can build IGSTK and IGSTKSandbox on Windows if you follow these instructions: http://public.kitware.com/IGSTKWIKI/index.php/How_to_build_IGSTK. When you configure IGSTKSandbox, make sure you turn OFF of IGSTKSandbox_USE_FLTK and turn ON of IGSTKSandbox_USE_OpenIGTLink. > > 2- I am trying to build IGSTK. I think that I could do it properly, I > followed all the steps but, when building the IGSTK with VS it does > not compiled. However it did compile when building IGSTKSandbox. Is > this normal/ok? > IGSTKSandbox depends on IGSTK and OpenIGTLink (you turn on IGSTKSandbox_USE_OpenIGTLink). So need successfully build IGSTK and OpenIGTLink first. Here are the instructions to build OpenIGTLink library: http://www.na-mic.org/Wiki/index.php/OpenIGTLink/Library/Build > 3- Once the IGSTKSandbox has been compiled with VS, what am I suppose > to do in order to use IGSTK? > If your IGSTKSandbox is successfully compiled, you should find an application called OIGTLinkTrackerBroadcasting in your binary directory. Run the application with an xml configuration file to get tracking data from a tracking device. The xml file is device specific. You can find examples in IGSTKSandbox/Examples/ApplicationsCommon/TrackerConfiguration/configurationFileExamples/ > 4- I am following the tutorial Advanced Tutorial Image Guided Therapy > in Slicer3 > (http://wiki.na-mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf). > All the information until slide 54 are pretty clear, however on slide > 55 it says: "run the IGSTK test program...". My question is: where is > that program?? > I tried to use the following path that is shown on that slide: > > $ ./bin/igstkSandboxTests > igstkAuroraTrackerToolObserverToOpenIGTLinkRelayTest > localhost 18944 10000 0 1 > > but is does not work. > See 3. > 5- Is there any other module/program available to connect the Aurora > tracker with slicer3? > IGSTK connects Slicer3 through OpenIGTLink module. Here the steps to do the connection: 1. Run Slicer3. 2. In Slicer3 interface, locate the Module list, go to OpenIGTLink module. 3. In OpenIGTLink module, go to Connectors, then click Add, Server, and Active. Now Slicer3 is ready to receive tracking data. 4. Run OIGTLinkTrackerBroadcasting auroraConfig.xml. Where auroraConfig.xml is your configuration file for your aurora system. Let me know if you have any further questions to set up your system. Best, Haiying Liu From ralph.brecheisen at gmail.com Mon May 11 16:29:42 2009 From: ralph.brecheisen at gmail.com (Ralph Brecheisen) Date: Mon, 11 May 2009 22:29:42 +0200 Subject: [slicer-devel] "Error in startup script" after building Slicer3 on Windows 7 Message-ID: <368a82280905111329p73e4ad33v96c8748fe97ce1db@mail.gmail.com> Hi, After running Cygwin in Administrator mode and disabling User Account Control, Slicer3 seems to build ok. However, if I run the Slicer3.exe, there's an error in the startup script: can't read "::Slicer3_REAL": no such variable while executing "set ::LAUNCH(launchCommand) $::Slicer3_REAL" (procedure "launch_RunProgram" line 42) invoked from within "launch_RunProgram" (file "C:/Applications/Slicer3-build/lib/Slicer3/launch.tcl" line 766) invoked from within "source $end(Slicer3_HOME)/lib/Slicer3/launch.tcl" (file "C:/Applications/Slicer3-build/Slicer3.exe/main.tcl" line 26) Any ideas how I can fix this? Thanks! Ralph -- "I respectfully decline the invitation to join your hallucination." - Scott Adams -- Ir. Ralph Brecheisen Biomedical Image Analysis @ Technical University of Eindhoven, Den Dolech 2, 5600 MB, Eindhoven, Netherlands Image-Guided Neurosurgery @ Maastricht University Hospital P. Debyelaan 25, 6229 HX, Maastricht, Netherlands Tel: +31-40-2475571 (Eindhoven) Tel: +31-43-3876292 (Maastricht) E-mail: r.brecheisen at tue.nl, ralph.brecheisen at gmail.com -------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/d533aaac/attachment.html From pieper at bwh.harvard.edu Mon May 11 16:39:32 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 11 May 2009 16:39:32 -0400 Subject: [slicer-devel] "Error in startup script" after building Slicer3 on Windows 7 In-Reply-To: <368a82280905111329p73e4ad33v96c8748fe97ce1db@mail.gmail.com> References: <368a82280905111329p73e4ad33v96c8748fe97ce1db@mail.gmail.com> Message-ID: <4A088D04.7010205@bwh.harvard.edu> Hi Ralph - There should be a file called Slicer3-real.exe in your build directory (probably in Slicer3-build/bin/Debug/Slicer3-real.exe). If this doesn't exist, then the build failed for some reason. If that's the case, then the thing to do is open Slicer3-build/Slicer3.sln in visual studio and try building the Slicer3-real target and see what errors you get... -Steve Ralph Brecheisen wrote: > Hi, > > After running Cygwin in Administrator mode and disabling User Account > Control, Slicer3 seems to build ok. > However, if I run the Slicer3.exe, there's an error in the startup script: > > can't read "::Slicer3_REAL": no such variable > while executing > "set ::LAUNCH(launchCommand) $::Slicer3_REAL" > (procedure "launch_RunProgram" line 42) > invoked from within > "launch_RunProgram" > (file "C:/Applications/Slicer3-build/lib/Slicer3/launch.tcl" line 766) > invoked from within > "source $end(Slicer3_HOME)/lib/Slicer3/launch.tcl" > (file "C:/Applications/Slicer3-build/Slicer3.exe/main.tcl" line 26) > > Any ideas how I can fix this? > Thanks! > > Ralph > > > > -- > "I respectfully decline the invitation to join your hallucination." > - Scott Adams > > -- > Ir. Ralph Brecheisen > > Biomedical Image Analysis @ Technical University of Eindhoven, Den > Dolech 2, 5600 MB, Eindhoven, Netherlands > Image-Guided Neurosurgery @ Maastricht University Hospital > P. Debyelaan 25, 6229 HX, Maastricht, Netherlands > > Tel: +31-40-2475571 (Eindhoven) > Tel: +31-43-3876292 (Maastricht) > E-mail: r.brecheisen at tue.nl , > ralph.brecheisen at gmail.com > -------------------------------------------------------------------------------------------------- > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From pieper at bwh.harvard.edu Mon May 11 16:47:40 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Mon, 11 May 2009 16:47:40 -0400 Subject: [slicer-devel] Compiling Slicer In-Reply-To: <81869fe90905111141v6a5a45b5v93e936677624fc6d@mail.gmail.com> References: <81869fe90905111136y6105b59di1dbc6edd7752d3fc@mail.gmail.com> <81869fe90905111141v6a5a45b5v93e936677624fc6d@mail.gmail.com> Message-ID: <4A088EEC.5060401@bwh.harvard.edu> Hi Raquel - What happens if you open the Slicer3.sln file in VCExpress and run the build from there? -Steve Raquel Torres Peralta wrote: > > Hello everyone. > > I?m trying to compile the 3D Slicer. I?ve solved some errors but now the > process stops with no more info of what?s happening so I don?t know what > to do. > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} Slicer3.SLN /build Debug > > > > child process exited abnormally > > > > Here?s what I have done so far for solving previous errors: > > - Run the compilation process using the administrator account > and disable the ?Turn User Account Control on or off". > > - I?ve added the next line to the slicer_variables.tcl file in > Slicer3 and also in slicer_variables.tcl file in Slicer3_2 to specify > the location of the building tool: > > > > if { [file exists "C:/Program Files (x86)/Microsoft Visual > Studio 9.0/Common7/IDE/VCExpress.exe"] } { set ::GENERATOR "Visual > Studio 9 2008" > > set ::MAKE "C:/Program Files (x86)/Microsoft Visual > Studio 9.0/Common7/IDE/VCExpress.exe" > > set ::COMPILER_PATH "C:/Program Files (x86)/Microsoft Visual > Studio 9.0/VC/bin" > > > > Next I got the error: > > CMake Error: The source "D:/cygwin/home/Raquel/Slicer3/CMakeLists.txt" > does not match the source > "D:/cygwin/home/Raquel/Slicer-3-2/CMakeLists.txt" used to generate > cache. Re-run cmake with a different source directory. > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpress.exe} Slicer3.SLN /build Debug > > > > child process exited abnormally > > > > Then I erased the CMakeCache.txt in my building directory and tried > again. Is then when I got the error: > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} Slicer3.SLN /build Debug ?Here it stops for a while - > > > > child process exited abnormally > > > > > > My OS: Windows Vista > > Compiler: Microsoft Visual C++ 2008 Express Edition > > CMake 2.7 > > > > Do you know what?s happening? > > Best regards, > > Raquel > > > > Here?s a copy of the messages: > > > > $ ./Slicer3/Scripts/getbuildtest.tcl -t "" > > getbuildtest: setting build list to > > Sourcing D:/cygwin/home/Raquel/Slicer3/slicer_variables.tcl > > Slicer3_HOME is D:/cygwin/home/Raquel/Slicer3 > > making with C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExp > > ress.exe > > running: svn switch http://svn.slicer.org/Slicer3/branches/Slicer-3-2 > > At revision 9444. > > > > running: sh ./Scripts/genlib.tcl > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib > > Stripped args = D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib > > Sourcing D:/cygwin/home/Raquel/Slicer3/slicer_variables.tcl > > running: svn co > http://svn.slicer.org/Slicer3-lib-mirrors/trunk/Binaries/Windows > > /python-build python-build > > Checked out revision 128. > > > > running: cvs -d :pserver:anoncvs:@www.kwwidgets.org:/cvsroot/KWWidgets login > > Logging in to :pserver:anoncvs at www.kwwidgets.org:2401/cvsroot/KWWidgets > > > > > running: cvs -q -z3 -d > :pserver:anoncvs at www.kwwidgets.org:/cvsroot/KWWidgets che > > ckout -r Slicer-3-2 KWWidgets > > > > running: > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe > > {-GVisual Studio 9 2008} > -DVTK_DIR:PATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3 > > -lib/VTK-build {-DCMAKE_CXX_COMPILER:STRING=C:/Program Files > (x86)/Microsoft Vis > > ual Studio 9.0/VC/bin/cl} > {-DCMAKE_CXX_COMPILER_FULLPATH:FILEPATH=C:/Program Fil > > es (x86)/Microsoft Visual Studio 9.0/VC/bin/cl} > -DBUILD_SHARED_LIBS:BOOL=ON -DCM > > AKE_SKIP_RPATH:BOOL=ON -DBUILD_EXAMPLES:BOOL=OFF > -DKWWidgets_BUILD_EXAMPLES:BOOL > > =OFF -DBUILD_TESTING:BOOL=OFF -DKWWidgets_BUILD_TESTING:BOOL=OFF > -DCMAKE_BUILD_T > > YPE:STRING=Debug ../KWWidgets > > -- Configuring done > > -- Generating done > > -- Build files have been written to: > D:/cygwin/home/Raquel/Slicer3-lib/KWWidgets > > -build > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} KWWidgets.SLN /build Debug > > > > All packages compiled. > > > > Slicer3_HOME is D:/cygwin/home/Raquel/Slicer3 > > cvs checkout: closing down connection to www.kwwidgets.org > > > child process exited abnormally > > running: > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe > > {-GVisual Studio 9 2008} {-DMAKECOMMAND:STRING=C:/Program Files > (x86)/Microsoft > > Visual Studio 9.0/Common7/IDE/VCExpress.exe} > {-DCMAKE_CXX_COMPILER:STRING=C:/Pro > > gram Files (x86)/Microsoft Visual Studio 9.0/VC/bin/cl} > {-DCMAKE_CXX_COMPILER_FU > > LLPATH:FILEPATH=C:/Program Files (x86)/Microsoft Visual Studio > 9.0/VC/bin/cl} -D > > ITK_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/Insight-build > -DKW > > Widgets_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/KWWidgets-buil > > d > -DTEEM_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/teem-build > -D > > IGSTK_DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/IGSTK-build > -DIN > > CR_TCL_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/tcl-build/l > > ib/lib/itcl3.2/itcl32.lib > -DINCR_TK_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slice > > r3/../Slicer3-lib/tcl-build/lib/lib/itk3.2/itk32.lib > -DSlicer3_USE_PYTHON=OFF -D > > PYTHON_INCLUDE_PATH:PATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/python-bui > > ld/include > -DPYTHON_LIBRARY:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-li > > b/python-build/Libs/python25.lib > -DSandBox_DIR:FILEPATH=D:/cygwin/home/Raquel/Sl > > icer3/../Slicer3-lib/NAMICSandBox -DCMAKE_BUILD_TYPE=Debug > -DSlicer3_VERSION_PAT > > CH:STRING=2009-05-10 -DCPACK_GENERATOR:STRING=NSIS > -DCPACK_PACKAGE_FILE_NAME:STR > > ING=Slicer3-3.2.2009-05-10-win32 -DSlicer3_USE_IGSTK=OFF > -DSlicer3_USE_NAVITRACK > > =OFF -DNAVITRACK_LIB_DIR:FILEPATH= -DNAVITRACK_INC_DIR:FILEPATH= > -DSLICERLIBCURL > > _DIR:FILEPATH=D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/cmcurl-build > -DCMAKE_ > > VERBOSE_MAKEFILE:BOOL=OFF D:/cygwin/home/Raquel/Slicer3 > > -- Configuring done > > -- Generating done > > -- Build files have been written to: D:/cygwin/home/Raquel/Slicer3-build > > > > running: > D:/cygwin/home/Raquel/Slicer3/../Slicer3-lib/CMake-build/bin/cmake.exe > > {-DMAKECOMMAND:STRING=C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7 > > /IDE/VCExpress.exe Slicer3.sln /build Debug /project ALL_BUILD} > D:/cygwin/home/R > > aquel/Slicer3 > > -- Configuring done > > -- Generating done > > -- Build files have been written to: D:/cygwin/home/Raquel/Slicer3-build > > > > running: {C:/Program Files (x86)/Microsoft Visual Studio > 9.0/Common7/IDE/VCExpre > > ss.exe} Slicer3.SLN /build Debug > > > > child process exited abnormally > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From julien.finet at kitware.com Mon May 11 17:38:07 2009 From: julien.finet at kitware.com (Julien Finet) Date: Mon, 11 May 2009 17:38:07 -0400 Subject: [slicer-devel] Register Images batch make module In-Reply-To: References: Message-ID: <5b9fa8230905111438m3bb85e38qed6bdf48e3135876@mail.gmail.com> Hi Harish, Make sure the "Register Images.bmm" file is generated in Slicer3-build/bin. Check the XML field Path, it must be set to the correct Register Images executable path. On my machine: "/home/julien/work/Slicer3-build/lib/Slicer3/Plugins/RegisterImages". Make sure you are running RegisterImagesBatchMake module in local mode, not condor. The option is in the BatchMake subsection of the RegisterImagesBatchMake user interface. Regards, Julien. On Mon, May 11, 2009 at 1:55 PM, Harish Doddi wrote: > > > Hi all, > > I am trying to use the registerd batch make module to find the optimum > parameters for registering a moving image to a fixed image. I have my image > datasets in .mha format. I have give the path for the fixed image file and > moving image directory and I gave moving image mask as *.mha. > > For Resample Image, I created a directory and gave that directory as input. > For Resample Image mask I gave I give the input as > Registered-${affineMaxIterations} since I want to find the resulting image > for each possible no of iterations ranging from affineIterations_min to > affineIterations_max in steps of affineIterations_step. For eg if I give min > value as 50 and max value 100 in steps of 10, I want the resulting image for > 50, 60, 70, 80, 90, 100.. > > But I dont understand why I get the following error in the output. > > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > > Kindly guide me if I am missing something. > > > Thanks, > Harish > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/fa97e775/attachment-0001.html From dpace at imaging.robarts.ca Mon May 11 22:06:30 2009 From: dpace at imaging.robarts.ca (Danielle Pace) Date: Mon, 11 May 2009 22:06:30 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> Message-ID: <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> Hi Abelardo, Can you please post the CMake options that you have at the moment? They should be equivalent to those at http://wiki.na-mic.org/Wiki/index.php/IGT:ToolKit/Build-IGSTK. I'm assuming that you've successfully built Slicer from source - is that correct? At first glance, it looks like there might be something wrong with ITK - perhaps ITK_DIR is set incorrectly? -Danielle -- Danielle Pace, BCmpH Masters Candidate, Biomedical Engineering Imaging Research Laboratories, Robarts Research Institute The University of Western Ontario, London, Ontario, Canada _____ From: Abelardo Escoto [mailto:aescoto2 at uwo.ca] Sent: May 11, 2009 12:10 PM To: Danielle Pace Cc: 'Developer, Slicer'; 'Junichi Tokuda'; 'Haiying Liu' Subject: Re: RE: [slicer-devel] Aurora tracker module help Hello Danielle: Thank you for you quick answer. After I use CMake to build IGSTK, I use VS to build the solution. The errors that VS are giving me are in the attached filed (image). Any idea how I can correct those errors? After compiling the IGSTKSandbox (which compiles properly), the folder igstkSandboxTests does not exist at the address: bin/igstkSandboxTests In the folder bin the only thing that I have is a folder named debug, and inside that one I do not have the igstkSandboxTests... Now I am wondering if I do not have that folder due to incorrectly selection of options during the CMake configuration...would it be possible?? Thank you for your time, Abelardo Escoto ----- Original Message ----- From: Danielle Pace Date: Monday, May 11, 2009 11:51 am Subject: RE: [slicer-devel] Aurora tracker module help To: 'Abelardo Escoto' , "'Developer, Slicer'" Cc: 'Junichi Tokuda' , 'Haiying Liu' > Hi Abelardo, > > 1) I don't think we have a Windows binary for IGSTK Sandbox, so you would have to build IGSTK and IGSTK Sandbox from source. > 2) If IGSTK didn't build, then you should resolve it - can you post the error messages? > 3) For interfacing with the Aurora tracking system, once you have IGSTK and IGSTKSandbox built correctly then you just need to run igstkAuroraTrackerToolObserverToOpenIGTLinkRelayTest to interface the Aurora with 3D Slicer - this test program is the code that uses IGSTK to enable tracking within Slicer. > 4) If you navigate to the IGSTKSandbox-build directory, where you built IGSTKSandbox, the IGSTK test program will be located in bin/igstkSandboxTests. > 5) I'm not aware of any, but perhaps Junichi or Haiying would be able to comment on if there have been any updates since the advanced navigation tutorial was made? > > Hope that helps, > > Danielle > > > -- > Danielle Pace, BCmpH > Masters Candidate, Biomedical Engineering > Imaging Research Laboratories, Robarts Research Institute > The University of Western Ontario, London, Ontario, Canada _____ > From: slicer-devel-bounces at bwh.harvard.edu [mailto:slicer-devel-bounces at bwh.harvard.edu] On Behalf Of Abelardo Escoto > Sent: May 11, 2009 11:35 AM > To: Developer, Slicer > Subject: [slicer-devel] Aurora tracker module help > > Hello: > > This is Abelardo Escoto, I am a Research Assistant working at CSTAR. Currently we are trying to develop some application using slicer3D, reason what I have been learning about it. > > I have some questions related with the IGT:Navigation-with-Aurora (http://wiki.na-mic.org/Wiki/index.php/IGT:ToolKit/Navigation-with-Aurora). > 1- In Step 3 it says that we could use the binaries, however, there is no binary for windows, OS that I am using. Is there any binary available for windows? > > 2- I am trying to build IGSTK. I think that I could do it properly, I followed all the steps but, when building the IGSTK with VS it does not compiled. However it did compile when building IGSTKSandbox. Is this normal/ok? > > 3- Once the IGSTKSandbox has been compiled with VS, what am I suppose to do in order to use IGSTK? > > 4- I am following the tutorial Advanced Tutorial Image Guided Therapy in Slicer3 (http://wiki.na-mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf). All the information until slide 54 are pretty clear, however on slide 55 it says: "run the IGSTK test program...". My question is: where is that program?? > I tried to use the following path that is shown on that slide: > > $ ./bin/igstkSandboxTests igstkAuroraTrackerToolObserverToOpenIGTLinkRelayTest > localhost 18944 10000 0 1 > > but is does not work. > > 5- Is there any other module/program available to connect the Aurora tracker with slicer3? > > Thank you very much for your time and attention, > > Abelardo Escoto -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090511/9d37b584/attachment-0001.html From fedorov at bwh.harvard.edu Tue May 12 08:55:23 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 12 May 2009 08:55:23 -0400 Subject: [slicer-devel] Register image parameters In-Reply-To: <68a07c2d0905111212k7a08c2d4m748dd75dd011d730@mail.gmail.com> References: <68a07c2d0905111212k7a08c2d4m748dd75dd011d730@mail.gmail.com> Message-ID: <82f536810905120555k7706d779n948fc7d64d0862bc@mail.gmail.com> Harish, Here's a relevant wiki page on tuning registration, which might be helpful to you: http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization AF On Mon, May 11, 2009 at 3:12 PM, Stephen Aylward wrote: > You can increase the sampling...but it may not help. ? Really depends > upon your data (e.g., sparseness of features wrt the metric being > used) as well as the transform (e.g., dimensionality). ? ?As well as > what you mean by "good results" ?:) > > The ITK Software Guide has a great chapter on image registration. > Covers topics like transforms, optimizers, parameter selection, etc. > ?You can get a free copy online at: > http://www.itk.org/ItkSoftwareGuide.pdf > > As a follow-on book, I suggest Insight Into Images book: > http://www.amazon.com/Insight-into-Images-Segmentation-Registration/dp/1568812175 > > Otherwise, perhaps post more info on the specific problems you are > having and perhaps we can provide some suggestions. > > Stephen > > On Mon, May 11, 2009 at 1:58 PM, Harish Doddi wrote: >> >> >> Hi all, >> >> I understand from the documentation that the parameters rigid sampling >> ratio, affine sampling ratio etc denote "the Portion of the image pixels to >> be used when computing the metric" >> >> So does this value range from 0-1. I am not getting good results for my >> dataset with default values. So can I give some higher values like 0.6, 0.8 >> etc or are there any side effects because of this. >> >> >> Thanks, >> Harish >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> > > > > -- > Stephen R. Aylward, Ph.D. > Director of Medical Imaging > Kitware, Inc. - North Carolina Office > http://www.kitware.com > (518) 371-3971 x300 > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From pieper at bwh.harvard.edu Tue May 12 09:00:51 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 12 May 2009 09:00:51 -0400 Subject: [slicer-devel] [slicer-users] Register image parameters In-Reply-To: <82f536810905120555k7706d779n948fc7d64d0862bc@mail.gmail.com> References: <68a07c2d0905111212k7a08c2d4m748dd75dd011d730@mail.gmail.com> <82f536810905120555k7706d779n948fc7d64d0862bc@mail.gmail.com> Message-ID: <4A097303.4040109@bwh.harvard.edu> Thanks for pointing that out, Andriy. I added a link off the module documentation: http://www.slicer.org/slicerWiki/index.php/Modules:RegisterImages-Documentation-3.4#Parameter_Choices Andriy Fedorov wrote: > Harish, > > Here's a relevant wiki page on tuning registration, which might be > helpful to you: > > http://www.na-mic.org/Wiki/index.php/ITK_Registration_Optimization > > AF > > > On Mon, May 11, 2009 at 3:12 PM, Stephen Aylward > wrote: >> You can increase the sampling...but it may not help. Really depends >> upon your data (e.g., sparseness of features wrt the metric being >> used) as well as the transform (e.g., dimensionality). As well as >> what you mean by "good results" :) >> >> The ITK Software Guide has a great chapter on image registration. >> Covers topics like transforms, optimizers, parameter selection, etc. >> You can get a free copy online at: >> http://www.itk.org/ItkSoftwareGuide.pdf >> >> As a follow-on book, I suggest Insight Into Images book: >> http://www.amazon.com/Insight-into-Images-Segmentation-Registration/dp/1568812175 >> >> Otherwise, perhaps post more info on the specific problems you are >> having and perhaps we can provide some suggestions. >> >> Stephen >> >> On Mon, May 11, 2009 at 1:58 PM, Harish Doddi wrote: >>> >>> Hi all, >>> >>> I understand from the documentation that the parameters rigid sampling >>> ratio, affine sampling ratio etc denote "the Portion of the image pixels to >>> be used when computing the metric" >>> >>> So does this value range from 0-1. I am not getting good results for my >>> dataset with default values. So can I give some higher values like 0.6, 0.8 >>> etc or are there any side effects because of this. >>> >>> >>> Thanks, >>> Harish >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject >>> >> >> >> -- >> Stephen R. Aylward, Ph.D. >> Director of Medical Imaging >> Kitware, Inc. - North Carolina Office >> http://www.kitware.com >> (518) 371-3971 x300 >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From stephen.aylward at kitware.com Tue May 12 09:12:47 2009 From: stephen.aylward at kitware.com (Stephen Aylward) Date: Tue, 12 May 2009 09:12:47 -0400 Subject: [slicer-devel] Reg Register Images Landmark Parameter In-Reply-To: References: Message-ID: <68a07c2d0905120612n5e62fb3eu394c242ad8564c7@mail.gmail.com> Yes - you can give those points. Make sure you then use a Pipelined registration method so that the initialization step is applied as part of the registration process. s On Mon, May 11, 2009 at 2:43 PM, Harish Doddi wrote: > Hi all, > > I am trying to use the register image module with landmarks as parameter. I > see the tutorial says landmark as " A vector string (comma separated base-3 > list) of the indexes of corresponding points in the fixed and moving images" > > From what I understand, I am trying to plot fiducial points on the fixed > image and moving image so that one point maps to other. Can I give these > fiducial points as input to these landmarks ie if in fixed image I have one > point as [-15.314, 2.0, 4.5] and moving image as [-10.0, 4, 3.5] Can I give > both these points as input to fixed/moving landmarks respectively or is > there something I am missing? > > > Thanks, > Harish > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe as the subject > -- Stephen R. Aylward, Ph.D. Director of Medical Imaging Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 From pieper at bwh.harvard.edu Tue May 12 09:21:47 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 12 May 2009 09:21:47 -0400 Subject: [slicer-devel] Reg Register Images Landmark Parameter In-Reply-To: <68a07c2d0905120612n5e62fb3eu394c242ad8564c7@mail.gmail.com> References: <68a07c2d0905120612n5e62fb3eu394c242ad8564c7@mail.gmail.com> Message-ID: <4A0977EB.4030305@bwh.harvard.edu> Hi Stephen - What's the syntax of the point list? Should Harish include the square brackets? -Steve Stephen Aylward wrote: > Yes - you can give those points. > > Make sure you then use a Pipelined registration method so that the > initialization step is applied as part of the registration process. > > s > > On Mon, May 11, 2009 at 2:43 PM, Harish Doddi wrote: >> Hi all, >> >> I am trying to use the register image module with landmarks as parameter. I >> see the tutorial says landmark as " A vector string (comma separated base-3 >> list) of the indexes of corresponding points in the fixed and moving images" >> >> From what I understand, I am trying to plot fiducial points on the fixed >> image and moving image so that one point maps to other. Can I give these >> fiducial points as input to these landmarks ie if in fixed image I have one >> point as [-15.314, 2.0, 4.5] and moving image as [-10.0, 4, 3.5] Can I give >> both these points as input to fixed/moving landmarks respectively or is >> there something I am missing? >> >> >> Thanks, >> Harish >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> > > > From stephen.aylward at kitware.com Tue May 12 09:26:23 2009 From: stephen.aylward at kitware.com (Stephen Aylward) Date: Tue, 12 May 2009 09:26:23 -0400 Subject: [slicer-devel] Reg Register Images Landmark Parameter In-Reply-To: <4A0977EB.4030305@bwh.harvard.edu> References: <68a07c2d0905120612n5e62fb3eu394c242ad8564c7@mail.gmail.com> <4A0977EB.4030305@bwh.harvard.edu> Message-ID: <68a07c2d0905120626w223c09bfm77f938b9c4be5e07@mail.gmail.com> Hi, No - simply comma separated, in triples x1,y1,z1,x2,y2,z2,... I need to update to support fiducial directly... s On Tue, May 12, 2009 at 9:21 AM, Steve Pieper wrote: > Hi Stephen - > > What's the syntax of the point list? ?Should Harish include the square > brackets? > > -Steve > > Stephen Aylward wrote: >> >> Yes - you can give those points. >> >> Make sure you then use a Pipelined registration method so that the >> initialization step is applied as part of the registration process. >> >> s >> >> On Mon, May 11, 2009 at 2:43 PM, Harish Doddi >> wrote: >>> >>> Hi all, >>> >>> I am trying to use the register image module with landmarks as parameter. >>> I >>> see the tutorial says landmark as " A vector string (comma separated >>> base-3 >>> list) of the indexes of corresponding points in the fixed and moving >>> images" >>> >>> From what I understand, I am trying to plot fiducial points on the fixed >>> image and moving image so that one point maps to other. Can I give these >>> fiducial points as input to these landmarks ie if in fixed image I have >>> one >>> point as [-15.314, 2.0, 4.5] and moving image as [-10.0, 4, 3.5] Can I >>> give >>> both these points as input to fixed/moving landmarks respectively or is >>> there something I am missing? >>> >>> >>> Thanks, >>> Harish >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu >>> with unsubscribe as the subject >>> >> >> >> > -- Stephen R. Aylward, Ph.D. Director of Medical Imaging Kitware, Inc. - North Carolina Office http://www.kitware.com (518) 371-3971 x300 From nrannou at bwh.harvard.edu Tue May 12 10:04:28 2009 From: nrannou at bwh.harvard.edu (Nicolas Rannou) Date: Tue, 12 May 2009 10:04:28 -0400 Subject: [slicer-devel] [slice-devel] Get the coordinate of a slice, in the red slice layout In-Reply-To: <82f536810905081336s10e1fecetbb2643dd583741a8@mail.gmail.com> References: <4A047885.9020901@bwh.harvard.edu> <82f536810905081336s10e1fecetbb2643dd583741a8@mail.gmail.com> Message-ID: <4A0981EC.8050808@bwh.harvard.edu> Thanks Andriy, looking at what you told me, I think thats not what I am trying to do. You told me how to get the coordinates of the cursor (mouse), in the red slice. What I want to do is to get the coordinates of the red Slice, in the loaded volume. Please find enclosed a screenshot to show you which one I'm trying to access. Thanks, Nicolas Andriy Fedorov wrote: > Nicolas, > > You can use this to transform from XY slice coordinates to the RAS space: > > vtkSlicerSliceGUI *sliceGUI = vtkSlicerApplicationGUI::SafeDownCast( > this->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red"); > > vtkMatrix4x4 *matrix = sliceGUI->GetLogic()->GetSliceNode()->GetXYToRAS(); > > hth > > AF > > > On Fri, May 8, 2009 at 2:23 PM, Nicolas Rannou wrote: > >> Hi all, >> >> for my application, I load a volume, then, I would like to get the >> coordinates of the slice I can see in the red slice layout, to extract >> this slice from my volume. >> I assume it must be something like that: >> >> /---------------------------------------------------------------------/ >> ->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red")->...? >> /---------------------------------------------------------------------/ >> >> but I can't find the exact way to get it. >> Does somebody know how I can access to it? >> >> Thanks, >> Nicolas >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject >> >> -------------- next part -------------- A non-text attachment was scrubbed... Name: slice_location.png Type: image/png Size: 82134 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/d559734d/attachment-0001.png From fedorov at bwh.harvard.edu Tue May 12 10:17:07 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 12 May 2009 10:17:07 -0400 Subject: [slicer-devel] [slice-devel] Get the coordinate of a slice, in the red slice layout In-Reply-To: <4A0981EC.8050808@bwh.harvard.edu> References: <4A047885.9020901@bwh.harvard.edu> <82f536810905081336s10e1fecetbb2643dd583741a8@mail.gmail.com> <4A0981EC.8050808@bwh.harvard.edu> Message-ID: <82f536810905120717r146155b5y88bec763e0beb2b4@mail.gmail.com> Nicolas, the code I sent you allows to transform coordinates of the cursor into coordinates in RAS space, which you can use to recover the slice coordinate. I agree, this may not be the most graceful way to do what you need. I do not know what is that most graceful way, but how about this one: vtkSlicerApplicationGUI *applicationGUI = this->GetGUI()->GetApplicationGUI(); slice = double(applicationGUI->GetMainSliceGUI("Red")->GetSliceController()->GetOffsetScale()->GetValue()); It is not easy to find, and I don't know this by heart. I just find the pieces of code where I remember this is done, and follow the example. Let us know if this resolves your question Fedorov On Tue, May 12, 2009 at 10:04 AM, Nicolas Rannou wrote: > Thanks Andriy, > > looking at what you told me, I think thats not what I am trying to do. > You told me how to get the coordinates of the cursor (mouse), in the red > slice. > > What I want to do is to get the coordinates of the red Slice, in the loaded > volume. > Please find enclosed a screenshot to show you which one I'm trying to > access. > > Thanks, > Nicolas > > Andriy Fedorov wrote: >> >> Nicolas, >> >> You can use this to transform from XY slice coordinates to the RAS space: >> >> vtkSlicerSliceGUI *sliceGUI = vtkSlicerApplicationGUI::SafeDownCast( >> ? ? ?this->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red"); >> >> vtkMatrix4x4 *matrix = sliceGUI->GetLogic()->GetSliceNode()->GetXYToRAS(); >> >> hth >> >> AF >> >> >> On Fri, May 8, 2009 at 2:23 PM, Nicolas Rannou >> wrote: >> >>> >>> Hi all, >>> >>> for my application, I load a volume, then, I would like to get the >>> coordinates of the slice I can see in the red slice layout, to extract >>> this slice from my volume. >>> I assume it must be something like that: >>> >>> /---------------------------------------------------------------------/ >>> ->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red")->...? >>> /---------------------------------------------------------------------/ >>> >>> but I can't find the exact way to get it. >>> Does somebody know how I can access to it? >>> >>> Thanks, >>> Nicolas >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >>> >>> > > From nrannou at bwh.harvard.edu Tue May 12 10:33:58 2009 From: nrannou at bwh.harvard.edu (Nicolas Rannou) Date: Tue, 12 May 2009 10:33:58 -0400 Subject: [slicer-devel] Integrating ITK filter in Slicer loadable module Message-ID: <4A0988D6.8060808@bwh.harvard.edu> Hi all, I'm currently trying to integrate an ITK Filter (N3MRIBiasFieldCorrectionImageFilter) to my Slicer3 loadable module. I'm trying, following the examples already implemented without success. /"The current 'best practice' for simple filters, per Steve Pieper and Jim Miller, is to encapsulate the ITK code inside a VTK class as done in Libs/vtkITK/vtkITKWandImageFilter.h and Libs/vtkITK/vtkITKWandImageFilter.cxx" /For my process, I have to integrate 4 filters. This encapsulating method looks good for me because from what I saw in the vtkITKWandImageFilter, I just have to add my itk filters in a method and it should work but it didn't so far. What surprises me is that this vtkITKWandImageFilter is integrated in a different way than the vtkITKGradientFilter. (please find the code enclosed) So I have some questions: -For a 'complex' filter, what is the best way for integration, the vtkITKWandImageFilter or the vtkITKGradientFilter? -Could I use an ITK filter in my loadable module just converting my input and output images from vtk to itk? Thanks, Nicolas -------------- next part -------------- A non-text attachment was scrubbed... Name: vtkITKGradientAnisotropicDiffusionImageFilter.cxx Type: text/x-c++src Size: 815 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/b7c34d8b/attachment.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: vtkITKGradientAnisotropicDiffusionImageFilter.h Type: text/x-chdr Size: 2826 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/b7c34d8b/attachment-0001.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: vtkITKWandImageFilter.cxx Type: text/x-c++src Size: 5023 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/b7c34d8b/attachment-0002.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: vtkITKWandImageFilter.h Type: text/x-chdr Size: 1920 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/b7c34d8b/attachment-0003.bin From nrannou at bwh.harvard.edu Tue May 12 11:03:09 2009 From: nrannou at bwh.harvard.edu (Nicolas Rannou) Date: Tue, 12 May 2009 11:03:09 -0400 Subject: [slicer-devel] [slice-devel] Get the coordinate of a slice, in the red slice layout In-Reply-To: <82f536810905120717r146155b5y88bec763e0beb2b4@mail.gmail.com> References: <4A047885.9020901@bwh.harvard.edu> <82f536810905081336s10e1fecetbb2643dd583741a8@mail.gmail.com> <4A0981EC.8050808@bwh.harvard.edu> <82f536810905120717r146155b5y88bec763e0beb2b4@mail.gmail.com> Message-ID: <4A098FAD.6080600@bwh.harvard.edu> Thanks it works perfectly: (code from vtkChangeTrackerAnalysisStep.h) /---------------------------------------------------------------------------------------/ #include "vtkSlicerSliceControllerWidget.h" #include "vtkSlicerSlicesControlGUI.h" #include "vtkKWScale.h" ... vtkSlicerApplicationGUI *applicationGUI = this->GetGUI()->GetApplicationGUI(); double oldSliceSetting[3]; oldSliceSetting[0] = double(applicationGUI->GetMainSliceGUI("Red")->GetSliceController()->GetOffsetScale()->GetValue()); oldSliceSetting[1] = double(applicationGUI->GetMainSliceGUI("Yellow")->GetSliceController()->GetOffsetScale()->GetValue()); oldSliceSetting[2] = double(applicationGUI->GetMainSliceGUI("Green")->GetSliceController()->GetOffsetScale()->GetValue()); .... /---------------------------------------------------------------------------------------/ Nicolas Andriy Fedorov wrote: > Nicolas, > > the code I sent you allows to transform coordinates of the cursor into > coordinates in RAS space, which you can use to recover the slice > coordinate. I agree, this may not be the most graceful way to do what > you need. > > I do not know what is that most graceful way, but how about this one: > > vtkSlicerApplicationGUI *applicationGUI = this->GetGUI()->GetApplicationGUI(); > slice = double(applicationGUI->GetMainSliceGUI("Red")->GetSliceController()->GetOffsetScale()->GetValue()); > > It is not easy to find, and I don't know this by heart. I just find > the pieces of code where I remember this is done, and follow the > example. > > Let us know if this resolves your question > > Fedorov > > > On Tue, May 12, 2009 at 10:04 AM, Nicolas Rannou > wrote: > >> Thanks Andriy, >> >> looking at what you told me, I think thats not what I am trying to do. >> You told me how to get the coordinates of the cursor (mouse), in the red >> slice. >> >> What I want to do is to get the coordinates of the red Slice, in the loaded >> volume. >> Please find enclosed a screenshot to show you which one I'm trying to >> access. >> >> Thanks, >> Nicolas >> >> Andriy Fedorov wrote: >> >>> Nicolas, >>> >>> You can use this to transform from XY slice coordinates to the RAS space: >>> >>> vtkSlicerSliceGUI *sliceGUI = vtkSlicerApplicationGUI::SafeDownCast( >>> this->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red"); >>> >>> vtkMatrix4x4 *matrix = sliceGUI->GetLogic()->GetSliceNode()->GetXYToRAS(); >>> >>> hth >>> >>> AF >>> >>> >>> On Fri, May 8, 2009 at 2:23 PM, Nicolas Rannou >>> wrote: >>> >>> >>>> Hi all, >>>> >>>> for my application, I load a volume, then, I would like to get the >>>> coordinates of the slice I can see in the red slice layout, to extract >>>> this slice from my volume. >>>> I assume it must be something like that: >>>> >>>> /---------------------------------------------------------------------/ >>>> ->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red")->...? >>>> /---------------------------------------------------------------------/ >>>> >>>> but I can't find the exact way to get it. >>>> Does somebody know how I can access to it? >>>> >>>> Thanks, >>>> Nicolas >>>> _______________________________________________ >>>> slicer-devel mailing list >>>> slicer-devel at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>> To unsubscribe: send email to >>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>> subject >>>> >>>> >>>> >> From fedorov at bwh.harvard.edu Tue May 12 11:14:58 2009 From: fedorov at bwh.harvard.edu (Andriy Fedorov) Date: Tue, 12 May 2009 11:14:58 -0400 Subject: [slicer-devel] [slice-devel] Get the coordinate of a slice, in the red slice layout In-Reply-To: <4A098FAD.6080600@bwh.harvard.edu> References: <4A047885.9020901@bwh.harvard.edu> <82f536810905081336s10e1fecetbb2643dd583741a8@mail.gmail.com> <4A0981EC.8050808@bwh.harvard.edu> <82f536810905120717r146155b5y88bec763e0beb2b4@mail.gmail.com> <4A098FAD.6080600@bwh.harvard.edu> Message-ID: <82f536810905120814k3c323054h9be09cb357c3242e@mail.gmail.com> "Reuse-and-conquer" seems to be the key in Slicer programming. The only problem is that you can never know if the component you are reusing follows the proper Slicer programming practices. I am glad I could help. On Tue, May 12, 2009 at 11:03 AM, Nicolas Rannou wrote: > Thanks it works perfectly: > (code from vtkChangeTrackerAnalysisStep.h) > > /---------------------------------------------------------------------------------------/ > #include "vtkSlicerSliceControllerWidget.h" > #include "vtkSlicerSlicesControlGUI.h" > #include "vtkKWScale.h" > > ... > > ? ? vtkSlicerApplicationGUI *applicationGUI ? ? = > this->GetGUI()->GetApplicationGUI(); > > ? ? double oldSliceSetting[3]; > ? ? oldSliceSetting[0] = > double(applicationGUI->GetMainSliceGUI("Red")->GetSliceController()->GetOffsetScale()->GetValue()); > ? ? oldSliceSetting[1] = > double(applicationGUI->GetMainSliceGUI("Yellow")->GetSliceController()->GetOffsetScale()->GetValue()); > ? ? oldSliceSetting[2] = > double(applicationGUI->GetMainSliceGUI("Green")->GetSliceController()->GetOffsetScale()->GetValue()); > > .... > /---------------------------------------------------------------------------------------/ > > Nicolas > Andriy Fedorov wrote: >> >> Nicolas, >> >> the code I sent you allows to transform coordinates of the cursor into >> coordinates in RAS space, which you can use to recover the slice >> coordinate. I agree, this may not be the most graceful way to do what >> you need. >> >> I do not know what is that most graceful way, but how about this one: >> >> vtkSlicerApplicationGUI *applicationGUI = >> this->GetGUI()->GetApplicationGUI(); >> slice = >> double(applicationGUI->GetMainSliceGUI("Red")->GetSliceController()->GetOffsetScale()->GetValue()); >> >> It is not easy to find, and I don't know this by heart. I just find >> the pieces of code where I remember this is done, and follow the >> example. >> >> Let us know if this resolves your question >> >> Fedorov >> >> >> On Tue, May 12, 2009 at 10:04 AM, Nicolas Rannou >> wrote: >> >>> >>> Thanks Andriy, >>> >>> looking at what you told me, I think thats not what I am trying to do. >>> You told me how to get the coordinates of the cursor (mouse), in the red >>> slice. >>> >>> What I want to do is to get the coordinates of the red Slice, in the >>> loaded >>> volume. >>> Please find enclosed a screenshot to show you which one I'm trying to >>> access. >>> >>> Thanks, >>> Nicolas >>> >>> Andriy Fedorov wrote: >>> >>>> >>>> Nicolas, >>>> >>>> You can use this to transform from XY slice coordinates to the RAS >>>> space: >>>> >>>> vtkSlicerSliceGUI *sliceGUI = vtkSlicerApplicationGUI::SafeDownCast( >>>> ? ? this->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red"); >>>> >>>> vtkMatrix4x4 *matrix = >>>> sliceGUI->GetLogic()->GetSliceNode()->GetXYToRAS(); >>>> >>>> hth >>>> >>>> AF >>>> >>>> >>>> On Fri, May 8, 2009 at 2:23 PM, Nicolas Rannou >>>> wrote: >>>> >>>> >>>>> >>>>> Hi all, >>>>> >>>>> for my application, I load a volume, then, I would like to get the >>>>> coordinates of the slice I can see in the red slice layout, to extract >>>>> this slice from my volume. >>>>> I assume it must be something like that: >>>>> >>>>> /---------------------------------------------------------------------/ >>>>> ->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red")->...? >>>>> /---------------------------------------------------------------------/ >>>>> >>>>> but I can't find the exact way to get it. >>>>> Does somebody know how I can access to it? >>>>> >>>>> Thanks, >>>>> Nicolas >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>>> subject >>>>> >>>>> >>>>> >>> >>> > > From pieper at bwh.harvard.edu Tue May 12 11:15:20 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 12 May 2009 11:15:20 -0400 Subject: [slicer-devel] [slice-devel] Get the coordinate of a slice, in the red slice layout In-Reply-To: <4A098FAD.6080600@bwh.harvard.edu> References: <4A047885.9020901@bwh.harvard.edu> <82f536810905081336s10e1fecetbb2643dd583741a8@mail.gmail.com> <4A0981EC.8050808@bwh.harvard.edu> <82f536810905120717r146155b5y88bec763e0beb2b4@mail.gmail.com> <4A098FAD.6080600@bwh.harvard.edu> Message-ID: <4A099288.7070207@bwh.harvard.edu> Hi Nicolas - Sounds like what you have will work. As Andriy said, it's probably not so graceful to pull the value from the widget. The underlying code that is used to get/set the values in the widget is in double vtkSlicerSliceLogic::GetSliceOffset() and void vtkSlicerSliceLogic::SetSliceOffset(double offset) To get between screen coordinates and voxel coordinates in slicer, the vtkMRMLSliceNode stores the XYToRAS matrix (i.e. how the slice plane is positioned in RAS space) and the vtkMRMLVolumeNode has the IJKToRAS matrix. Best, Steve Nicolas Rannou wrote: > Thanks it works perfectly: > (code from vtkChangeTrackerAnalysisStep.h) > > /---------------------------------------------------------------------------------------/ > #include "vtkSlicerSliceControllerWidget.h" > #include "vtkSlicerSlicesControlGUI.h" > #include "vtkKWScale.h" > > ... > > vtkSlicerApplicationGUI *applicationGUI = this->GetGUI()->GetApplicationGUI(); > > double oldSliceSetting[3]; > oldSliceSetting[0] = double(applicationGUI->GetMainSliceGUI("Red")->GetSliceController()->GetOffsetScale()->GetValue()); > oldSliceSetting[1] = double(applicationGUI->GetMainSliceGUI("Yellow")->GetSliceController()->GetOffsetScale()->GetValue()); > oldSliceSetting[2] = double(applicationGUI->GetMainSliceGUI("Green")->GetSliceController()->GetOffsetScale()->GetValue()); > > .... > /---------------------------------------------------------------------------------------/ > > Nicolas > Andriy Fedorov wrote: >> Nicolas, >> >> the code I sent you allows to transform coordinates of the cursor into >> coordinates in RAS space, which you can use to recover the slice >> coordinate. I agree, this may not be the most graceful way to do what >> you need. >> >> I do not know what is that most graceful way, but how about this one: >> >> vtkSlicerApplicationGUI *applicationGUI = this->GetGUI()->GetApplicationGUI(); >> slice = double(applicationGUI->GetMainSliceGUI("Red")->GetSliceController()->GetOffsetScale()->GetValue()); >> >> It is not easy to find, and I don't know this by heart. I just find >> the pieces of code where I remember this is done, and follow the >> example. >> >> Let us know if this resolves your question >> >> Fedorov >> >> >> On Tue, May 12, 2009 at 10:04 AM, Nicolas Rannou >> wrote: >> >>> Thanks Andriy, >>> >>> looking at what you told me, I think thats not what I am trying to do. >>> You told me how to get the coordinates of the cursor (mouse), in the red >>> slice. >>> >>> What I want to do is to get the coordinates of the red Slice, in the loaded >>> volume. >>> Please find enclosed a screenshot to show you which one I'm trying to >>> access. >>> >>> Thanks, >>> Nicolas >>> >>> Andriy Fedorov wrote: >>> >>>> Nicolas, >>>> >>>> You can use this to transform from XY slice coordinates to the RAS space: >>>> >>>> vtkSlicerSliceGUI *sliceGUI = vtkSlicerApplicationGUI::SafeDownCast( >>>> this->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red"); >>>> >>>> vtkMatrix4x4 *matrix = sliceGUI->GetLogic()->GetSliceNode()->GetXYToRAS(); >>>> >>>> hth >>>> >>>> AF >>>> >>>> >>>> On Fri, May 8, 2009 at 2:23 PM, Nicolas Rannou >>>> wrote: >>>> >>>> >>>>> Hi all, >>>>> >>>>> for my application, I load a volume, then, I would like to get the >>>>> coordinates of the slice I can see in the red slice layout, to extract >>>>> this slice from my volume. >>>>> I assume it must be something like that: >>>>> >>>>> /---------------------------------------------------------------------/ >>>>> ->GetGUI()->GetApplicationGUI())->GetMainSliceGUI("Red")->...? >>>>> /---------------------------------------------------------------------/ >>>>> >>>>> but I can't find the exact way to get it. >>>>> Does somebody know how I can access to it? >>>>> >>>>> Thanks, >>>>> Nicolas >>>>> _______________________________________________ >>>>> slicer-devel mailing list >>>>> slicer-devel at bwh.harvard.edu >>>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>>>> To unsubscribe: send email to >>>>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>>>> subject >>>>> >>>>> >>>>> >>> > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From aescoto2 at uwo.ca Tue May 12 13:39:34 2009 From: aescoto2 at uwo.ca (Abelardo Escoto) Date: Tue, 12 May 2009 13:39:34 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> Message-ID: An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/c6b1ac91/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: CMake IGTKSandbox options.JPG Type: image/jpeg Size: 114166 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/c6b1ac91/attachment-0001.jpe From harish.slicer at gmail.com Tue May 12 14:15:08 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Tue, 12 May 2009 11:15:08 -0700 Subject: [slicer-devel] register images batch make Message-ID: Hi all, I just downloaded the binary of slicer 3.5 alpha and tried to run register images batch make module. I get an error (attached in the screenshot) when running this module. I guess that there is some part of the code that has been hardcoded.. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/f7432ad5/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: RegisterImageBatchMakeError.tiff Type: image/tiff Size: 248049 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/f7432ad5/attachment-0001.tiff From hliu at bwh.harvard.edu Tue May 12 14:56:08 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Tue, 12 May 2009 14:56:08 -0400 (EDT) Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> Message-ID: Hi Abelardo, Please see my text interleaved below. Haiying On Tue, 12 May 2009, Abelardo Escoto wrote: > Hello All!!! > > 1- Thank you Haiying for the information about: > > "Application called? OIGTLinkTrackerBroadcasting" > > I found these application and the xml configuration file. However, when I double click the OIGTLinkTrackerBroadcasting.exe, the following msg appears and I can not > run it: > > This application has failed to start because ITKCommond.dll was not found. Re-installing the applications may fix this problem" > > What can I do in order to fix this? > You are welcome. You need set your PATH variable pointing to those ITK libraries. > > 2- The parameters that I am using for the xml configuration file, based on the example are as follow (is the format right??) > > > > ??? 20 > ??? > ????? 2 > ????? 115200 > ????? 7 > ????? N > ????? 1 > ????? 0 > ??? > > ??? > ????? Needle Placer > ????? 1 > ????? 0 > ????? C:\Program Files\Northern Digital Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465-5D.rom > ??? > > ??? > ????? US probe > ????? 1 > ????? 1 > ????? C:\Program Files\Northern Digital Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom > ????? > > > I put the file in the some folder where the OIGTLinkTrackerBroadcasting application is. > a. The xml file configures your tracking device such as Aurora. If your computer has a serial port, write your port number in the field "2", and let unchanged other fields of the serial communication. b. The sample xml file specified two probes. For your first testing case, you can just comment out the bottom one. c. Inside the tool tag, add this tag at the bottom: localhost:18994 where 18994 is port number Slicer is waiting for. if you run OIGTLinkTrackerBroadcasting and Slicer on different computers, replace "localhost" by the ip address of the Slicer computer. > > 3-? I am attaching you the CMake screen for the IGTKSandbox. It is important to clarify that the IGTKSandbox is compiling properly. > I think your compiling options should work. Let me know if you have more questions. Best, Haiying Liu From aescoto2 at uwo.ca Tue May 12 16:33:20 2009 From: aescoto2 at uwo.ca (Abelardo Escoto) Date: Tue, 12 May 2009 16:33:20 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> Message-ID: hello again!! Thank you again. I moved a lot of .dll to the folder System32. The application now opens/runs... My question now is: now what?? I can not see the console because it opens and closes right away. Can I still follow the tutorial and I can not follow the tutorial??? http://wiki.na-mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf??? even though I am not using the same application?? If I can, how do I connect it with Slicer?? Sorry for my lack of knowledge and thank you very much for your time, Abelardo ----- Original Message ----- From: Haiying Liu Date: Tuesday, May 12, 2009 3:26 pm Subject: Re: RE: RE: [slicer-devel] Aurora tracker module help To: Abelardo Escoto Cc: Danielle Pace , "'Developer, Slicer'" > Hi Abelardo, > > Please see my text interleaved below. > > > Haiying > > On Tue, 12 May 2009, Abelardo Escoto wrote: > > > Hello All!!! > > > > 1- Thank you Haiying for the information about: > > > > "Application called? OIGTLinkTrackerBroadcasting" > > > > I found these application and the xml configuration file. > However, when I double click the > OIGTLinkTrackerBroadcasting.exe, the following msg appears and I > can not > > run it: > > > > This application has failed to start because ITKCommond.dll > was not found. Re-installing the applications may fix this problem" > > > > What can I do in order to fix this? > > > > You are welcome. You need set your PATH variable pointing to > those ITK > libraries. > > > > > > 2- The parameters that I am using for the xml configuration > file, based on the example are as follow (is the format right??) > > > > > > > > ??? 20 > > ??? > > ????? 2 > > ????? 115200 > > ????? 7 > > ????? N > > ????? 1 > > ????? 0 > > ??? > > > > ??? > > ????? Needle Placer > > ????? 1 > > ????? 0 > > ????? C:\Program Files\Northern Digital > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465-5D.rom > > ??? > > > > ??? > > ????? US probe > > ????? 1 > > ????? 1 > > ????? C:\Program Files\Northern Digital > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom> ????? > > > > > > I put the file in the some folder where the > OIGTLinkTrackerBroadcasting application is. > > > > a. > The xml file configures your tracking device such as Aurora. If > your > computer has a serial port, write your port number in the field > "2", and let unchanged other fields > of the serial > communication. > > b. > The sample xml file specified two probes. For your first testing > case, you > can just comment out the bottom one. > > c. Inside the tool tag, add this tag at the bottom: > ????? > localhost:18994 > where 18994 is port number Slicer is waiting for. > if you run? OIGTLinkTrackerBroadcasting and Slicer on > different computers, > replace "localhost" by the ip address of the Slicer computer. > > > > > 3-? I am attaching you the CMake screen for the IGTKSandbox. > It is important to clarify that the IGTKSandbox is compiling properly. > > > > I think your compiling options should work. > > > Let me know if you have more questions. > > Best, > > > > Haiying Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/2deff800/attachment.html From harish.slicer at gmail.com Tue May 12 19:03:45 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Tue, 12 May 2009 16:03:45 -0700 Subject: [slicer-devel] Register Images batch make module In-Reply-To: References: Message-ID: This Problem is fixed.. Thanks Julien for your help. Harish On Mon, May 11, 2009 at 10:55 AM, Harish Doddi wrote: > > > Hi all, > > I am trying to use the registerd batch make module to find the optimum > parameters for registering a moving image to a fixed image. I have my image > datasets in .mha format. I have give the path for the fixed image file and > moving image directory and I gave moving image mask as *.mha. > > For Resample Image, I created a directory and gave that directory as input. > For Resample Image mask I gave I give the input as > Registered-${affineMaxIterations} since I want to find the resulting image > for each possible no of iterations ranging from affineIterations_min to > affineIterations_max in steps of affineIterations_step. For eg if I give min > value as 50 and max value 100 in steps of 10, I want the resulting image for > 50, 60, 70, 80, 90, 100.. > > But I dont understand why I get the following error in the output. > > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > Error: Could not run RegisterImagesPath-NOTFOUND: > No such file or directory > > Kindly guide me if I am missing something. > > > Thanks, > Harish > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/f10d5cd9/attachment.html From harish.slicer at gmail.com Tue May 12 19:32:58 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Tue, 12 May 2009 16:32:58 -0700 Subject: [slicer-devel] Python Surface ICP Registration Message-ID: Hi All, We are looking into the python icp registration module developed by Luca Antiga and Daniel Blezek and tried to register two femur models of two patients. The results with affine registration look promising. We would like to extend the functionality of this module to include MR/CT images as target surface. We have an initial idea of how we should go about doing this. I see the code SurfaceICPRegistration.py in the directory Slicer3/Modules/Python. But I know that Slicer has a bug when building with python modules. So I have been turning the python option off while building slicer. As a result, my slicer executable doesnt have python modules. Could someone please guide me in understanding how to add/compile the code, integrating into slicer and also provide some links where I can use python with ITK, VTK. Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090512/a590c9da/attachment-0001.html From dpace at imaging.robarts.ca Wed May 13 10:28:14 2009 From: dpace at imaging.robarts.ca (Danielle Pace) Date: Wed, 13 May 2009 10:28:14 -0400 Subject: [slicer-devel] Real-time frame grabbing in Slicer Message-ID: Hi everyone, Our lab relies heavily on real-time imaging (for example, ultrasound and laparoscopy) for surgical guidance during various forms of interventions, which we have previously accomplished quite easily using Matrox frame grabbers and vtkMILVideoSource. I was wondering if Slicer also had a module that implements this functionality - from my search online and through the Slicer source code, it doesn't seem like it does. I understand that the OpenIGTLink protocol could be used to interface to a small IGSTK program using IGSTK's frame grabbing functionality - is this the recommended approach? I wonder if it's overkill to have two programs running for something so simple as video grabbing, and whether this would introduce lag into the system. Thanks, Danielle -- Danielle Pace, BCmpH Masters Candidate, Biomedical Engineering Imaging Research Laboratories, Robarts Research Institute The University of Western Ontario, London, Ontario, Canada -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090513/0b29eebf/attachment.html From hliu at bwh.harvard.edu Wed May 13 10:40:11 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Wed, 13 May 2009 10:40:11 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> Message-ID: <4A0ADBCB.8030101@bwh.harvard.edu> Hi Abelardo, Now you have Slicer and IGSTK built on your computer. Here are the steps to setup the communication between Slicer and Aurora: 1) Run Slicer 2) In Slicer, go Modules -> OpenIGTLinkIF -> Connectors -> Add -> Server -> Active. Slicer is now ready to receive tracking data 3) Connect Aurora to your computer through a serial port 4) Customize your config file for your Aurora (the xml file we talked about yesterday) 5) Run OIGTLinkTrackerBroadcasting.exe auroraConfig.xml 6) In Slicer, go Modules -> Transform. If everything works ok, you should see tracking data dynamically change as you move your Aurora probe around (within its tracking scope) Let me know if that works for you, Haiying Abelardo Escoto wrote: > hello again!! > Thank you again. I moved a lot of .dll to the folder System32. The > application now opens/runs... > > My question now is: now what?? I can not see the console because it > opens and closes right away. > Can I still follow the tutorial and I can not follow the tutorial > http://wiki.na-mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf > even though I am not using the same application?? If I can, how do I > connect it with Slicer?? > > Sorry for my lack of knowledge and thank you very much for your time, > Abelardo > > ----- Original Message ----- > From: Haiying Liu > Date: Tuesday, May 12, 2009 3:26 pm > Subject: Re: RE: RE: [slicer-devel] Aurora tracker module help > To: Abelardo Escoto > Cc: Danielle Pace , "'Developer, Slicer'" > > > > Hi Abelardo, > > > > Please see my text interleaved below. > > > > > > Haiying > > > > On Tue, 12 May 2009, Abelardo Escoto wrote: > > > > > Hello All!!! > > > > > > 1- Thank you Haiying for the information about: > > > > > > "Application called OIGTLinkTrackerBroadcasting" > > > > > > I found these application and the xml configuration file. > > However, when I double click the > > OIGTLinkTrackerBroadcasting.exe, the following msg appears and I > > can not > > > run it: > > > > > > This application has failed to start because ITKCommond.dll > > was not found. Re-installing the applications may fix this problem" > > > > > > What can I do in order to fix this? > > > > > > > You are welcome. You need set your PATH variable pointing to > > those ITK > > libraries. > > > > > > > > > > 2- The parameters that I am using for the xml configuration > > file, based on the example are as follow (is the format right??) > > > > > > > > > > > > 20 > > > > > > 2 > > > 115200 > > > 7 > > > N > > > 1 > > > 0 > > > > > > > > > > > > Needle Placer > > > 1 > > > 0 > > > C:\Program Files\Northern Digital > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465-5D.rom > > > > > > > > > > > > US probe > > > 1 > > > 1 > > > C:\Program Files\Northern Digital > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom> > > > > > > > > > > I put the file in the some folder where the > > OIGTLinkTrackerBroadcasting application is. > > > > > > > a. > > The xml file configures your tracking device such as Aurora. If > > your > > computer has a serial port, write your port number in the field > > "2", and let unchanged other fields > > of the serial > > communication. > > > > b. > > The sample xml file specified two probes. For your first testing > > case, you > > can just comment out the bottom one. > > > > c. Inside the tool tag, add this tag at the bottom: > > > > localhost:18994 > > where 18994 is port number Slicer is waiting for. > > if you run OIGTLinkTrackerBroadcasting and Slicer on > > different computers, > > replace "localhost" by the ip address of the Slicer computer. > > > > > > > > 3- I am attaching you the CMake screen for the IGTKSandbox. > > It is important to clarify that the IGTKSandbox is compiling properly. > > > > > > > I think your compiling options should work. > > > > > > Let me know if you have more questions. > > > > Best, > > > > > > > > Haiying Liu From aescoto2 at uwo.ca Wed May 13 13:55:19 2009 From: aescoto2 at uwo.ca (Abelardo Escoto) Date: Wed, 13 May 2009 13:55:19 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: <4A0ADBCB.8030101@bwh.harvard.edu> References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> <4A0ADBCB.8030101@bwh.harvard.edu> Message-ID: Hello: The information was great...it kind of worked out. What is happening is this: 1- The command OIGTLinkTrackerBroadcasting.exe auroraConfig.xml works well, it opens the port and the Aurora tracker bips. 2- One of the problems is that there is nothing in the transforms module. Same situation happens on the data module. If I create a new transform node still nothing happens. 3- This is the configuration file that I am using: ? ??? 20 ??? ????? 1 ????? 115200 ????? 7 ????? N ????? 1 ????? 0 ??? ??? ????? Needle Placer ????? 1 ????? 0 ????? C:\Program Files\Northern Digital Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)??? \8700465-5D.rom ????? localhost:18994 ???? ? ???? ?????? ? ? ? US probe ????? 2 ????? C:\Program Files\Northern Digital Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom ????? localhost:18944 ???? ?????? I am wondering if something is wrong with it. I have removed the srom files but still does not work. 4- Does slicer display two transformation matrices? each for one tool attached to the aurora?? Thank you for your time, Abelardo ----- Original Message ----- From: Haiying Liu Date: Wednesday, May 13, 2009 10:39 am Subject: Re: [slicer-devel] Aurora tracker module help To: Abelardo Escoto Cc: Danielle Pace , "'Developer, Slicer'" > Hi Abelardo, > > Now you have Slicer and IGSTK built on your computer. Here are > the steps > to setup the communication between Slicer and Aurora: > > 1) Run Slicer > 2) In Slicer, go Modules -> OpenIGTLinkIF -> Connectors -> Add - > > Server > -> Active. Slicer is now ready to receive tracking data > 3) Connect Aurora to your computer through a serial port > 4) Customize your config file for your Aurora (the xml file we > talked > about yesterday) > 5) Run > OIGTLinkTrackerBroadcasting.exe auroraConfig.xml > 6) In Slicer, go Modules -> Transform. If everything works ok, > you > should see tracking data dynamically change as you move your > Aurora > probe around (within its tracking scope) > > Let me know if that works for you, > > > Haiying > > Abelardo Escoto wrote: > > hello again!! > > Thank you again. I moved a lot of .dll to the folder System32. > The > > application now opens/runs... > > > > My question now is: now what?? I can not see the console > because it > > opens and closes right away. > > Can I still follow the tutorial and I can not follow the > tutorial??? > > http://wiki.na- > mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf??? > > even though I am not using the same application?? If I can, > how do I > > connect it with Slicer?? > > > > Sorry for my lack of knowledge and thank you very much for > your time, > > Abelardo > > > > ----- Original Message ----- > > From: Haiying Liu > > Date: Tuesday, May 12, 2009 3:26 pm > > Subject: Re: RE: RE: [slicer-devel] Aurora tracker module help > > To: Abelardo Escoto > > Cc: Danielle Pace , "'Developer, > Slicer'" > > > > > > > Hi Abelardo, > > > > > > Please see my text interleaved below. > > > > > > > > > Haiying > > > > > > On Tue, 12 May 2009, Abelardo Escoto wrote: > > > > > > > Hello All!!! > > > > > > > > 1- Thank you Haiying for the information about: > > > > > > > > "Application called? OIGTLinkTrackerBroadcasting" > > > > > > > > I found these application and the xml configuration file. > > > However, when I double click the > > > OIGTLinkTrackerBroadcasting.exe, the following msg appears > and I > > > can not > > > > run it: > > > > > > > > This application has failed to start because ITKCommond.dll > > > was not found. Re-installing the applications may fix this > problem"> > > > > > > What can I do in order to fix this? > > > > > > > > > > You are welcome. You need set your PATH variable pointing to > > > those ITK > > > libraries. > > > > > > > > > > > > > > 2- The parameters that I am using for the xml configuration > > > file, based on the example are as follow (is the format right??) > > > > > > > > > > > > > > > >???? 20 > > > >???? > > > >?????? 2 > > > >?????? > 115200> > > >?????? 7 > > > >?????? N > > > >?????? > 1> > > >?????? 0 > > > >???? > > > > > > > >???? > > > >?????? Needle > Placer > > > >?????? > 1> > > >?????? 0 > > > >?????? > C:\Program Files\Northern Digital > > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465- > 5D.rom> > >???? > > > > > > > >???? > > > >?????? US probe > > > >?????? > 1> > > >?????? 1 > > > >?????? > C:\Program Files\Northern Digital > > > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom> > >?????? > > > > > > > > > > > > I put the file in the some folder where the > > > OIGTLinkTrackerBroadcasting application is. > > > > > > > > > > a. > > > The xml file configures your tracking device such as Aurora. If > > > your > > > computer has a serial port, write your port number in the field > > > "2", and let unchanged other fields > > > of the serial > > > communication. > > > > > > b. > > > The sample xml file specified two probes. For your first testing > > > case, you > > > can just comment out the bottom one. > > > > > > c. Inside the tool tag, add this tag at the bottom: > > >????? > > > localhost:18994 > > > where 18994 is port number Slicer is waiting for. > > > if you run? OIGTLinkTrackerBroadcasting and Slicer on > > > different computers, > > > replace "localhost" by the ip address of the Slicer computer. > > > > > > > > > > > 3-? I am attaching you the CMake screen for the > IGTKSandbox.> > It is important to clarify that the IGTKSandbox > is compiling properly. > > > > > > > > > > I think your compiling options should work. > > > > > > > > > Let me know if you have more questions. > > > > > > Best, > > > > > > > > > > > > Haiying Liu > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090513/863a9470/attachment-0001.html From hliu at bwh.harvard.edu Wed May 13 14:13:37 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Wed, 13 May 2009 14:13:37 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> <4A0ADBCB.8030101@bwh.harvard.edu> Message-ID: <4A0B0DD1.9030903@bwh.harvard.edu> Hi Abelardo, Please check my text below. Abelardo Escoto wrote: > Hello: > > The information was great...it kind of worked out. > What is happening is this: > > 1- The command OIGTLinkTrackerBroadcasting.exe auroraConfig.xml works > well, it opens the port and the Aurora tracker bips. > Did you see the Aurora light turn from yellow to green? > 2- One of the problems is that there is nothing in the transforms > module. Same situation happens on the data module. If I create a new > transform node still nothing happens. > If everything works, a transform node will be created automatically in Slicer data module, which will be visible when you open Transform module. The node name will depend on the tool name specified in the xml file. Please check the OpenIGTLinkIF module. Did you add a connector first? After you ran OIGTLinkTrackerBroadcasting.exe, did you see the connector status changed from Waiting to ON? If it's still Waiting, that means communication between Slicer and Aurora is not setup yet. > 3- This is the configuration file that I am using: > > > > > > 20 > > 1 > 115200 > 7 > N > 1 > 0 > > > > Needle Placer > 1 > 0 > C:\Program Files\Northern Digital > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001) \8700465-5D.rom > localhost:18994 > > > > US probe > 2 > C:\Program Files\Northern Digital > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom > localhost:18944 > > > > > > I am wondering if something is wrong with it. > I have removed the srom files but still does not work. > > To simplify your testing, use only ONE probe first. Remove the second tool from your xml file. Insert your probe into the first port on Aurora. > 4- Does slicer display two transformation matrices? each for one tool > attached to the aurora?? > You can have multiple probes and so you can have multiple transform nodes created in Slicer. > Thank you for your time, > Abelardo > > ----- Original Message ----- > From: Haiying Liu > Date: Wednesday, May 13, 2009 10:39 am > Subject: Re: [slicer-devel] Aurora tracker module help > To: Abelardo Escoto > Cc: Danielle Pace , "'Developer, Slicer'" > > > > Hi Abelardo, > > > > Now you have Slicer and IGSTK built on your computer. Here are > > the steps > > to setup the communication between Slicer and Aurora: > > > > 1) Run Slicer > > 2) In Slicer, go Modules -> OpenIGTLinkIF -> Connectors -> Add - > > > Server > > -> Active. Slicer is now ready to receive tracking data > > 3) Connect Aurora to your computer through a serial port > > 4) Customize your config file for your Aurora (the xml file we > > talked > > about yesterday) > > 5) Run > > OIGTLinkTrackerBroadcasting.exe auroraConfig.xml > > 6) In Slicer, go Modules -> Transform. If everything works ok, > > you > > should see tracking data dynamically change as you move your > > Aurora > > probe around (within its tracking scope) > > > > Let me know if that works for you, > > > > > > Haiying > > > > Abelardo Escoto wrote: > > > hello again!! > > > Thank you again. I moved a lot of .dll to the folder System32. > > The > > > application now opens/runs... > > > > > > My question now is: now what?? I can not see the console > > because it > > > opens and closes right away. > > > Can I still follow the tutorial and I can not follow the > > tutorial > > > http://wiki.na- > > mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf > > > even though I am not using the same application?? If I can, > > how do I > > > connect it with Slicer?? > > > > > > Sorry for my lack of knowledge and thank you very much for > > your time, > > > Abelardo > > > > > > ----- Original Message ----- > > > From: Haiying Liu > > > Date: Tuesday, May 12, 2009 3:26 pm > > > Subject: Re: RE: RE: [slicer-devel] Aurora tracker module help > > > To: Abelardo Escoto > > > Cc: Danielle Pace , "'Developer, > > Slicer'" > > > > > > > > > > Hi Abelardo, > > > > > > > > Please see my text interleaved below. > > > > > > > > > > > > Haiying > > > > > > > > On Tue, 12 May 2009, Abelardo Escoto wrote: > > > > > > > > > Hello All!!! > > > > > > > > > > 1- Thank you Haiying for the information about: > > > > > > > > > > "Application called OIGTLinkTrackerBroadcasting" > > > > > > > > > > I found these application and the xml configuration file. > > > > However, when I double click the > > > > OIGTLinkTrackerBroadcasting.exe, the following msg appears > > and I > > > > can not > > > > > run it: > > > > > > > > > > This application has failed to start because ITKCommond.dll > > > > was not found. Re-installing the applications may fix this > > problem"> > > > > > > > What can I do in order to fix this? > > > > > > > > > > > > > You are welcome. You need set your PATH variable pointing to > > > > those ITK > > > > libraries. > > > > > > > > > > > > > > > > > > 2- The parameters that I am using for the xml configuration > > > > file, based on the example are as follow (is the format right??) > > > > > > > > > > > > > > > > > > > > 20 > > > > > > > > > > 2 > > > > > > > 115200> > > > > 7 > > > > > N > > > > > > > 1> > > > > 0 > > > > > > > > > > > > > > > > > > > > Needle > > Placer > > > > > > > 1> > > > > 0 > > > > > > > C:\Program Files\Northern Digital > > > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465- > > 5D.rom> > > > > > > > > > > > > > > > > > US probe > > > > > > > 1> > > > > 1 > > > > > > > C:\Program Files\Northern Digital > > > > > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom> > > > > > > > > > > > > > > > > > > I put the file in the some folder where the > > > > OIGTLinkTrackerBroadcasting application is. > > > > > > > > > > > > > a. > > > > The xml file configures your tracking device such as Aurora. If > > > > your > > > > computer has a serial port, write your port number in the field > > > > "2", and let unchanged other fields > > > > of the serial > > > > communication. > > > > > > > > b. > > > > The sample xml file specified two probes. For your first testing > > > > case, you > > > > can just comment out the bottom one. > > > > > > > > c. Inside the tool tag, add this tag at the bottom: > > > > > > > > localhost:18994 > > > > where 18994 is port number Slicer is waiting for. > > > > if you run OIGTLinkTrackerBroadcasting and Slicer on > > > > different computers, > > > > replace "localhost" by the ip address of the Slicer computer. > > > > > > > > > > > > > > 3- I am attaching you the CMake screen for the > > IGTKSandbox.> > It is important to clarify that the IGTKSandbox > > is compiling properly. > > > > > > > > > > > > > I think your compiling options should work. > > > > > > > > > > > > Let me know if you have more questions. > > > > > > > > Best, > > > > > > > > > > > > > > > > Haiying Liu > > From aescoto2 at uwo.ca Wed May 13 14:16:15 2009 From: aescoto2 at uwo.ca (Abelardo Escoto) Date: Wed, 13 May 2009 14:16:15 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: <4A0ADBCB.8030101@bwh.harvard.edu> References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> <4A0ADBCB.8030101@bwh.harvard.edu> Message-ID: This is me again... Correction on my last email. 1- I can only use one tool at a time, cygwin dos not allow me to send information to the same host. It gives me this msg: Tools 'Needle Placer' and 'US probe' send data to the same destination. 2- The xml file that I am using is this ? ??? 20 ??? ????? 1 ????? 115200 ????? 7 ????? N ????? 1 ????? 0 ??? ??? ????? Needle Placer ????? 1 ????? 0 ????? C:\Program Files\Northern Digital Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465-5D.rom ????? localhost:18944 ??? Still does not work, neither the indicator moves when I move my tool. Thank you, Abelardo ----- Original Message ----- From: Haiying Liu Date: Wednesday, May 13, 2009 10:39 am Subject: Re: [slicer-devel] Aurora tracker module help To: Abelardo Escoto Cc: Danielle Pace , "'Developer, Slicer'" > Hi Abelardo, > > Now you have Slicer and IGSTK built on your computer. Here are > the steps > to setup the communication between Slicer and Aurora: > > 1) Run Slicer > 2) In Slicer, go Modules -> OpenIGTLinkIF -> Connectors -> Add - > > Server > -> Active. Slicer is now ready to receive tracking data > 3) Connect Aurora to your computer through a serial port > 4) Customize your config file for your Aurora (the xml file we > talked > about yesterday) > 5) Run > OIGTLinkTrackerBroadcasting.exe auroraConfig.xml > 6) In Slicer, go Modules -> Transform. If everything works ok, > you > should see tracking data dynamically change as you move your > Aurora > probe around (within its tracking scope) > > Let me know if that works for you, > > > Haiying > > Abelardo Escoto wrote: > > hello again!! > > Thank you again. I moved a lot of .dll to the folder System32. > The > > application now opens/runs... > > > > My question now is: now what?? I can not see the console > because it > > opens and closes right away. > > Can I still follow the tutorial and I can not follow the > tutorial??? > > http://wiki.na- > mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf??? > > even though I am not using the same application?? If I can, > how do I > > connect it with Slicer?? > > > > Sorry for my lack of knowledge and thank you very much for > your time, > > Abelardo > > > > ----- Original Message ----- > > From: Haiying Liu > > Date: Tuesday, May 12, 2009 3:26 pm > > Subject: Re: RE: RE: [slicer-devel] Aurora tracker module help > > To: Abelardo Escoto > > Cc: Danielle Pace , "'Developer, > Slicer'" > > > > > > > Hi Abelardo, > > > > > > Please see my text interleaved below. > > > > > > > > > Haiying > > > > > > On Tue, 12 May 2009, Abelardo Escoto wrote: > > > > > > > Hello All!!! > > > > > > > > 1- Thank you Haiying for the information about: > > > > > > > > "Application called? OIGTLinkTrackerBroadcasting" > > > > > > > > I found these application and the xml configuration file. > > > However, when I double click the > > > OIGTLinkTrackerBroadcasting.exe, the following msg appears > and I > > > can not > > > > run it: > > > > > > > > This application has failed to start because ITKCommond.dll > > > was not found. Re-installing the applications may fix this > problem"> > > > > > > What can I do in order to fix this? > > > > > > > > > > You are welcome. You need set your PATH variable pointing to > > > those ITK > > > libraries. > > > > > > > > > > > > > > 2- The parameters that I am using for the xml configuration > > > file, based on the example are as follow (is the format right??) > > > > > > > > > > > > > > > >???? 20 > > > >???? > > > >?????? 2 > > > >?????? > 115200> > > >?????? 7 > > > >?????? N > > > >?????? > 1> > > >?????? 0 > > > >???? > > > > > > > >???? > > > >?????? Needle > Placer > > > >?????? > 1> > > >?????? 0 > > > >?????? > C:\Program Files\Northern Digital > > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465- > 5D.rom> > >???? > > > > > > > >???? > > > >?????? US probe > > > >?????? > 1> > > >?????? 1 > > > >?????? > C:\Program Files\Northern Digital > > > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom> > >?????? > > > > > > > > > > > > I put the file in the some folder where the > > > OIGTLinkTrackerBroadcasting application is. > > > > > > > > > > a. > > > The xml file configures your tracking device such as Aurora. If > > > your > > > computer has a serial port, write your port number in the field > > > "2", and let unchanged other fields > > > of the serial > > > communication. > > > > > > b. > > > The sample xml file specified two probes. For your first testing > > > case, you > > > can just comment out the bottom one. > > > > > > c. Inside the tool tag, add this tag at the bottom: > > >????? > > > localhost:18994 > > > where 18994 is port number Slicer is waiting for. > > > if you run? OIGTLinkTrackerBroadcasting and Slicer on > > > different computers, > > > replace "localhost" by the ip address of the Slicer computer. > > > > > > > > > > > 3-? I am attaching you the CMake screen for the > IGTKSandbox.> > It is important to clarify that the IGTKSandbox > is compiling properly. > > > > > > > > > > I think your compiling options should work. > > > > > > > > > Let me know if you have more questions. > > > > > > Best, > > > > > > > > > > > > Haiying Liu > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090513/ba28afae/attachment-0001.html From aescoto2 at uwo.ca Wed May 13 14:19:19 2009 From: aescoto2 at uwo.ca (Abelardo Escoto) Date: Wed, 13 May 2009 14:19:19 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: <4A0B0DD1.9030903@bwh.harvard.edu> References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> <4A0ADBCB.8030101@bwh.harvard.edu> <4A0B0DD1.9030903@bwh.harvard.edu> Message-ID: Hi Haiying: Yes, it went from yellow to green. Yes, it went from waiting to on. No, it did not create automatically the transform node. Thank you!!!! ----- Original Message ----- From: Haiying Liu Date: Wednesday, May 13, 2009 2:13 pm Subject: Re: [slicer-devel] Aurora tracker module help To: Abelardo Escoto Cc: "'Developer, Slicer'" > Hi Abelardo, > > Please check my text below. > > Abelardo Escoto wrote: > > Hello: > > > > The information was great...it kind of worked out. > > What is happening is this: > > > > 1- The command OIGTLinkTrackerBroadcasting.exe > auroraConfig.xml works > > well, it opens the port and the Aurora tracker bips. > > > > Did you see the Aurora light turn from yellow to green? > > > 2- One of the problems is that there is nothing in the > transforms > > module. Same situation happens on the data module. If I create > a new > > transform node still nothing happens. > > > If everything works, a transform node will be created > automatically in > Slicer data module, which will be visible when you open > Transform > module. The node name will depend on the tool name specified in > the xml > file. > > Please check the OpenIGTLinkIF module. Did you add a connector > first? > After you ran OIGTLinkTrackerBroadcasting.exe, did you see the > connector > status changed from Waiting to ON? If it's still Waiting, that > means > communication between Slicer and Aurora is not setup yet. > > > 3- This is the configuration file that I am using: > > > > > >? > > > > > >???? 20 > >???? > >?????? 1 > >?????? > 115200>?????? 7 > >?????? N > >?????? 1 > >?????? 0 > >???? > > > >???? > >?????? Needle Placer > >?????? > 1>?????? 0 > >?????? C:\Program > Files\Northern Digital > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)??? > \8700465-5D.rom > >?????? > localhost:18994>????? > >? > >????? > ????? > >?????? US probe > >?????? > 2>?????? C:\Program Files\Northern Digital > > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom>?????? localhost:18944 > >????? > ????? > > > > > > > > > > I am wondering if something is wrong with it. > > I have removed the srom files but still does not work. > > > > > To simplify your testing, use only ONE probe first. Remove the > second > tool from your xml file. Insert your probe into the first port > on Aurora. > > > > 4- Does slicer display two transformation matrices? each for > one tool > > attached to the aurora?? > > > You can have multiple probes and so you can have multiple > transform > nodes created in Slicer. > > > Thank you for your time, > > Abelardo > > > > ----- Original Message ----- > > From: Haiying Liu > > Date: Wednesday, May 13, 2009 10:39 am > > Subject: Re: [slicer-devel] Aurora tracker module help > > To: Abelardo Escoto > > Cc: Danielle Pace , "'Developer, > Slicer'" > > > > > > > Hi Abelardo, > > > > > > Now you have Slicer and IGSTK built on your computer. Here are > > > the steps > > > to setup the communication between Slicer and Aurora: > > > > > > 1) Run Slicer > > > 2) In Slicer, go Modules -> OpenIGTLinkIF -> Connectors -> > Add - > > > > Server > > > -> Active. Slicer is now ready to receive tracking data > > > 3) Connect Aurora to your computer through a serial port > > > 4) Customize your config file for your Aurora (the xml file we > > > talked > > > about yesterday) > > > 5) Run > > > OIGTLinkTrackerBroadcasting.exe auroraConfig.xml > > > 6) In Slicer, go Modules -> Transform. If everything works ok, > > > you > > > should see tracking data dynamically change as you move your > > > Aurora > > > probe around (within its tracking scope) > > > > > > Let me know if that works for you, > > > > > > > > > Haiying > > > > > > Abelardo Escoto wrote: > > > > hello again!! > > > > Thank you again. I moved a lot of .dll to the folder System32. > > > The > > > > application now opens/runs... > > > > > > > > My question now is: now what?? I can not see the console > > > because it > > > > opens and closes right away. > > > > Can I still follow the tutorial and I can not follow the > > > tutorial?? > > > > http://wiki.na- > > > > mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf?? > > > > even though I am not using the same application?? If I can, > > > how do I > > > > connect it with Slicer?? > > > > > > > > Sorry for my lack of knowledge and thank you very much for > > > your time, > > > > Abelardo > > > > > > > > ----- Original Message ----- > > > > From: Haiying Liu > > > > Date: Tuesday, May 12, 2009 3:26 pm > > > > Subject: Re: RE: RE: [slicer-devel] Aurora tracker module help > > > > To: Abelardo Escoto > > > > Cc: Danielle Pace , "'Developer, > > > Slicer'" > > > > > > > > > > > > > Hi Abelardo, > > > > > > > > > > Please see my text interleaved below. > > > > > > > > > > > > > > > Haiying > > > > > > > > > > On Tue, 12 May 2009, Abelardo Escoto wrote: > > > > > > > > > > > Hello All!!! > > > > > > > > > > > > 1- Thank you Haiying for the information about: > > > > > > > > > > > > "Application called? OIGTLinkTrackerBroadcasting" > > > > > > > > > > > > I found these application and the xml configuration file. > > > > > However, when I double click the > > > > > OIGTLinkTrackerBroadcasting.exe, the following msg appears > > > and I > > > > > can not > > > > > > run it: > > > > > > > > > > > > This application has failed to start because ITKCommond.dll > > > > > was not found. Re-installing the applications may fix this > > > problem"> > > > > > > > > What can I do in order to fix this? > > > > > > > > > > > > > > > > You are welcome. You need set your PATH variable > pointing to > > > > > those ITK > > > > > libraries. > > > > > > > > > > > > > > > > > > > > > > 2- The parameters that I am using for the xml configuration > > > > > file, based on the example are as follow (is the format > right??)> > > > > > > > > > > > > > > > > > > > > > >???? 20 > > > > > >???? > > > > > > >?????? 2 > > > > > >????? > > > 115200> > > > > >?????? > 7> > > > > >?????? N > > > > > >????? > > > 1> > > > > >?????? > 0> > > > > >???? > > > > > > > > > > > >???? > > > > > >?????? Needle > > > Placer > > > > > >????? > > > 1> > > > > >?????? > 0> > > > > >????? > > > C:\Program Files\Northern Digital > > > > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465- > > > 5D.rom> > >???? > > > > > > > > > > > >???? > > > > > >?????? US probe > > > > > >????? > > > 1> > > > > >?????? > 1> > > > > >????? > > > C:\Program Files\Northern Digital > > > > > > > > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom>> > >?????? > > > > > > > > > > > > > > > > > > I put the file in the some folder where the > > > > > OIGTLinkTrackerBroadcasting application is. > > > > > > > > > > > > > > > > a. > > > > > The xml file configures your tracking device such as > Aurora. If > > > > > your > > > > > computer has a serial port, write your port number in > the field > > > > > "2", and let unchanged other fields > > > > > of the serial > > > > > communication. > > > > > > > > > > b. > > > > > The sample xml file specified two probes. For your first > testing> > > > case, you > > > > > can just comment out the bottom one. > > > > > > > > > > c. Inside the tool tag, add this tag at the bottom: > > > > >???? > > > > > localhost:18994 > > > > > where 18994 is port number Slicer is waiting for. > > > > > if you run? OIGTLinkTrackerBroadcasting and Slicer on > > > > > different computers, > > > > > replace "localhost" by the ip address of the Slicer computer. > > > > > > > > > > > > > > > > > 3-? I am attaching you the CMake screen for the > > > IGTKSandbox.> > It is important to clarify that the IGTKSandbox > > > is compiling properly. > > > > > > > > > > > > > > > > I think your compiling options should work. > > > > > > > > > > > > > > > Let me know if you have more questions. > > > > > > > > > > Best, > > > > > > > > > > > > > > > > > > > > Haiying Liu > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090513/32b50916/attachment-0001.html From hliu at bwh.harvard.edu Wed May 13 14:27:08 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Wed, 13 May 2009 14:27:08 -0400 Subject: [slicer-devel] Aurora tracker module help In-Reply-To: References: <5492C4CD96CE4A55ABA9AF6F295B6209@irus.robarts.ca> <138C73A46A2A4118B869833F9E736C54@irus.robarts.ca> <4A0ADBCB.8030101@bwh.harvard.edu> <4A0B0DD1.9030903@bwh.harvard.edu> Message-ID: <4A0B10FC.2040305@bwh.harvard.edu> Abelardo Escoto wrote: > Hi Haiying: > > Yes, it went from yellow to green. > Yes, it went from waiting to on. > No, it did not create automatically the transform node. I believe you are getting close to make it work. Move your probe around while making sure the probe is inside the tracking scope of your aurora; that is, make sure the green light on the probe stay stable instead of flashing. Then, check back Slicer in Data or Transform module to see if a Transform node is created. Haiying > > Thank you!!!! > > ----- Original Message ----- > From: Haiying Liu > Date: Wednesday, May 13, 2009 2:13 pm > Subject: Re: [slicer-devel] Aurora tracker module help > To: Abelardo Escoto > Cc: "'Developer, Slicer'" > > > Hi Abelardo, > > > > Please check my text below. > > > > Abelardo Escoto wrote: > > > Hello: > > > > > > The information was great...it kind of worked out. > > > What is happening is this: > > > > > > 1- The command OIGTLinkTrackerBroadcasting.exe > > auroraConfig.xml works > > > well, it opens the port and the Aurora tracker bips. > > > > > > > Did you see the Aurora light turn from yellow to green? > > > > > 2- One of the problems is that there is nothing in the > > transforms > > > module. Same situation happens on the data module. If I create > > a new > > > transform node still nothing happens. > > > > > If everything works, a transform node will be created > > automatically in > > Slicer data module, which will be visible when you open > > Transform > > module. The node name will depend on the tool name specified in > > the xml > > file. > > > > Please check the OpenIGTLinkIF module. Did you add a connector > > first? > > After you ran OIGTLinkTrackerBroadcasting.exe, did you see the > > connector > > status changed from Waiting to ON? If it's still Waiting, that > > means > > communication between Slicer and Aurora is not setup yet. > > > > > 3- This is the configuration file that I am using: > > > > > > > > > > > > > > > > > > 20 > > > > > > 1 > > > > > 115200> 7 > > > N > > > 1 > > > 0 > > > > > > > > > > > > Needle Placer > > > > > 1> > 0 > > > C:\Program > > Files\Northern Digital > > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001) > > \8700465-5D.rom > > > > > localhost:18994> > > > > > > > > > > > US probe > > > > > 2> C:\Program > Files\Northern Digital > > > > > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom> > localhost:18944 > > > > > > > > > > > > > > > > > > > > I am wondering if something is wrong with it. > > > I have removed the srom files but still does not work. > > > > > > > > To simplify your testing, use only ONE probe first. Remove the > > second > > tool from your xml file. Insert your probe into the first port > > on Aurora. > > > > > > > 4- Does slicer display two transformation matrices? each for > > one tool > > > attached to the aurora?? > > > > > You can have multiple probes and so you can have multiple > > transform > > nodes created in Slicer. > > > > > Thank you for your time, > > > Abelardo > > > > > > ----- Original Message ----- > > > From: Haiying Liu > > > Date: Wednesday, May 13, 2009 10:39 am > > > Subject: Re: [slicer-devel] Aurora tracker module help > > > To: Abelardo Escoto > > > Cc: Danielle Pace , "'Developer, > > Slicer'" > > > > > > > > > > Hi Abelardo, > > > > > > > > Now you have Slicer and IGSTK built on your computer. Here are > > > > the steps > > > > to setup the communication between Slicer and Aurora: > > > > > > > > 1) Run Slicer > > > > 2) In Slicer, go Modules -> OpenIGTLinkIF -> Connectors -> > > Add - > > > > > Server > > > > -> Active. Slicer is now ready to receive tracking data > > > > 3) Connect Aurora to your computer through a serial port > > > > 4) Customize your config file for your Aurora (the xml file we > > > > talked > > > > about yesterday) > > > > 5) Run > > > > OIGTLinkTrackerBroadcasting.exe auroraConfig.xml > > > > 6) In Slicer, go Modules -> Transform. If everything works ok, > > > > you > > > > should see tracking data dynamically change as you move your > > > > Aurora > > > > probe around (within its tracking scope) > > > > > > > > Let me know if that works for you, > > > > > > > > > > > > Haiying > > > > > > > > Abelardo Escoto wrote: > > > > > hello again!! > > > > > Thank you again. I moved a lot of .dll to the folder System32. > > > > The > > > > > application now opens/runs... > > > > > > > > > > My question now is: now what?? I can not see the console > > > > because it > > > > > opens and closes right away. > > > > > Can I still follow the tutorial and I can not follow the > > > > tutorial > > > > > http://wiki.na- > > > > > > mic.org/Wiki/images/f/f9/IGTAdvancedNavigation.pdf > > > > > even though I am not using the same application?? If I can, > > > > how do I > > > > > connect it with Slicer?? > > > > > > > > > > Sorry for my lack of knowledge and thank you very much for > > > > your time, > > > > > Abelardo > > > > > > > > > > ----- Original Message ----- > > > > > From: Haiying Liu > > > > > Date: Tuesday, May 12, 2009 3:26 pm > > > > > Subject: Re: RE: RE: [slicer-devel] Aurora tracker module help > > > > > To: Abelardo Escoto > > > > > Cc: Danielle Pace , "'Developer, > > > > Slicer'" > > > > > > > > > > > > > > > > Hi Abelardo, > > > > > > > > > > > > Please see my text interleaved below. > > > > > > > > > > > > > > > > > > Haiying > > > > > > > > > > > > On Tue, 12 May 2009, Abelardo Escoto wrote: > > > > > > > > > > > > > Hello All!!! > > > > > > > > > > > > > > 1- Thank you Haiying for the information about: > > > > > > > > > > > > > > "Application called OIGTLinkTrackerBroadcasting" > > > > > > > > > > > > > > I found these application and the xml configuration file. > > > > > > However, when I double click the > > > > > > OIGTLinkTrackerBroadcasting.exe, the following msg appears > > > > and I > > > > > > can not > > > > > > > run it: > > > > > > > > > > > > > > This application has failed to start because ITKCommond.dll > > > > > > was not found. Re-installing the applications may fix this > > > > problem"> > > > > > > > > > What can I do in order to fix this? > > > > > > > > > > > > > > > > > > > You are welcome. You need set your PATH variable > > pointing to > > > > > > those ITK > > > > > > libraries. > > > > > > > > > > > > > > > > > > > > > > > > > > 2- The parameters that I am using for the xml configuration > > > > > > file, based on the example are as follow (is the format > > right??)> > > > > > > > > > > > > > > > > > > > > > > > > > 20 > > > > > > > > > > > > > > > > 2 > > > > > > > > > > > 115200> > > > > > > > > 7> > > > > > > N > > > > > > > > > > > 1> > > > > > > > > 0> > > > > > > > > > > > > > > > > > > > > > > > > > > > Needle > > > > Placer > > > > > > > > > > > 1> > > > > > > > > 0> > > > > > > > > > > C:\Program Files\Northern Digital > > > > > > Inc\ToolBox\Aurora_5DOF_Sensor(8700465.001)\8700465- > > > > 5D.rom> > > > > > > > > > > > > > > > > > > > > > > > US probe > > > > > > > > > > > 1> > > > > > > > > 1> > > > > > > > > > > C:\Program Files\Northern Digital > > > > > > > > > > > > > Inc\ToolBox\Aurora_6DOF_Sensor(8700468.002)\8700468.rom>> > > > > > > > > > > > > > > > > > > > > > > > > I put the file in the some folder where the > > > > > > OIGTLinkTrackerBroadcasting application is. > > > > > > > > > > > > > > > > > > > a. > > > > > > The xml file configures your tracking device such as > > Aurora. If > > > > > > your > > > > > > computer has a serial port, write your port number in > > the field > > > > > > "2", and let unchanged other fields > > > > > > of the serial > > > > > > communication. > > > > > > > > > > > > b. > > > > > > The sample xml file specified two probes. For your first > > testing> > > > case, you > > > > > > can just comment out the bottom one. > > > > > > > > > > > > c. Inside the tool tag, add this tag at the bottom: > > > > > > > > > > > > localhost:18994 > > > > > > where 18994 is port number Slicer is waiting for. > > > > > > if you run OIGTLinkTrackerBroadcasting and Slicer on > > > > > > different computers, > > > > > > replace "localhost" by the ip address of the Slicer computer. > > > > > > > > > > > > > > > > > > > > 3- I am attaching you the CMake screen for the > > > > IGTKSandbox.> > It is important to clarify that the IGTKSandbox > > > > is compiling properly. > > > > > > > > > > > > > > > > > > > I think your compiling options should work. > > > > > > > > > > > > > > > > > > Let me know if you have more questions. > > > > > > > > > > > > Best, > > > > > > > > > > > > > > > > > > > > > > > > Haiying Liu > > > > > > From hliu at bwh.harvard.edu Wed May 13 23:20:17 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Wed, 13 May 2009 23:20:17 -0400 Subject: [slicer-devel] Real-time frame grabbing in Slicer In-Reply-To: References: Message-ID: <4A0B8DF1.6040109@bwh.harvard.edu> Hi Danielle, Let's check with Junichi about real-time imaging support in Slicer. Junichi, does ProstateNav module in Slicer support frame grabbing? Best, Haiying Danielle Pace wrote: > Hi everyone, > > Our lab relies heavily on real-time imaging (for example, ultrasound > and laparoscopy) for surgical guidance during various forms of > interventions, which we have previously accomplished quite easily > using Matrox frame grabbers and vtkMILVideoSource. I was wondering if > Slicer also had a module that implements this functionality ? from my > search online and through the Slicer source code, it doesn?t seem like > it does. I understand that the OpenIGTLink protocol could be used to > interface to a small IGSTK program using IGSTK?s frame grabbing > functionality ? is this the recommended approach? I wonder if it?s > overkill to have two programs running for something so simple as video > grabbing, and whether this would introduce lag into the system. > > > Thanks, > > Danielle > > -- > > Danielle Pace, BCmpH > > Masters Candidate, Biomedical Engineering > > Imaging Research Laboratories, Robarts Research Institute > > The University of Western Ontario, London, Ontario, Canada > >------------------------------------------------------------------------ > >_______________________________________________ >slicer-devel mailing list >slicer-devel at bwh.harvard.edu >http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject > From raquel.torresperalta at gmail.com Wed May 13 23:20:17 2009 From: raquel.torresperalta at gmail.com (Raquel Torres Peralta) Date: Wed, 13 May 2009 20:20:17 -0700 Subject: [slicer-devel] Compiling still... Message-ID: <81869fe90905132020s1c3c3799y3e86ea78b4e33962@mail.gmail.com> Hello everyone. I am now on a Win XP machine with MS Visual Studio 2005. I am building the 3.4 version. I get the next error when running getbuildtest: CMake Error: Error in cmake code at D:/cygwin/home/ADMINISTRADOR EHUI/Slicer/Slicer3/Applications/CLI/CMakeLists.txt :568: Parse error. Function missing ending ")". Instead found left paren with text " (". After that, I modified the file and make it run again. Now it does this: $ ./Slicer3/Scripts/getbuildtest.tcl Sourcing D:/cygwin/home/ADMINISTRADOR EHUI/Slicer/Slicer3/slicer_variables.tcl Slicer3_HOME is D:/cygwin/home/ADMINI~1/Slicer/Slicer3 making with c:/Program Files/Microsoft Visual Studio 8/Common7/IDE/devenv.exe running: svn switch http://svn.slicer.org/Slicer3/branches/Slicer-3-4 and seems to never come back or move forward. Any ideas? It will be better to build 3.2 instead? Best regards, Raquel -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090513/0186aff6/attachment.html From hayes at bwh.harvard.edu Wed May 13 23:31:17 2009 From: hayes at bwh.harvard.edu (Kathryn Hayes) Date: Wed, 13 May 2009 23:31:17 -0400 (EDT) Subject: [slicer-devel] Compiling still... In-Reply-To: <81869fe90905132020s1c3c3799y3e86ea78b4e33962@mail.gmail.com> References: <81869fe90905132020s1c3c3799y3e86ea78b4e33962@mail.gmail.com> Message-ID: Hi Racquel, Wendy was also having that problem earlier, but it was solved for her by running getbuildtest.tcl --update again. Could you try that and see if that works for you? Thanks, Katie On Wed, 13 May 2009, Raquel Torres Peralta wrote: > Hello everyone. > > I am now on a Win XP machine with MS Visual Studio 2005. I am building the > 3.4 version. > > I get the next error when running getbuildtest: > CMake Error: Error in cmake code at > D:/cygwin/home/ADMINISTRADOR > EHUI/Slicer/Slicer3/Applications/CLI/CMakeLists.txt > :568: > Parse error. Function missing ending ")". Instead found left paren with > text " > (". > > After that, I modified the file and make it run again. > > Now it does this: > > $ ./Slicer3/Scripts/getbuildtest.tcl > Sourcing D:/cygwin/home/ADMINISTRADOR > EHUI/Slicer/Slicer3/slicer_variables.tcl > Slicer3_HOME is D:/cygwin/home/ADMINI~1/Slicer/Slicer3 > making with c:/Program Files/Microsoft Visual Studio > 8/Common7/IDE/devenv.exe > running: svn switch http://svn.slicer.org/Slicer3/branches/Slicer-3-4 > and seems to never come back or move forward. Any ideas? It will be better > to build 3.2 instead? > > Best regards, > Raquel > From raquel.torresperalta at gmail.com Thu May 14 02:19:22 2009 From: raquel.torresperalta at gmail.com (Raquel Torres Peralta) Date: Wed, 13 May 2009 23:19:22 -0700 Subject: [slicer-devel] Compiling still... In-Reply-To: <81869fe90905132222o25176b42s21961da89c44175c@mail.gmail.com> References: <81869fe90905132020s1c3c3799y3e86ea78b4e33962@mail.gmail.com> <81869fe90905132102s84d8844ra85f1dfa826f990e@mail.gmail.com> <81869fe90905132132lc35f666u356a90783061499a@mail.gmail.com> <81869fe90905132222o25176b42s21961da89c44175c@mail.gmail.com> Message-ID: <81869fe90905132319i5258033dn4ca709c9d596cbc6@mail.gmail.com> Hi everyone. It's been an hour since the getbuildtest.tcl stopped in the next message and has not moved. Any suggestions? Best wishes, Raquel running: D:/cygwin/home/ADMINI~1/Slicer/Slicer3/../Slicer3-lib/CMake-build/bin/ test -D Experimental -C Debug Site: CHEWI Build name: Win32-vs8 Create new tag: 20090514-0444 - Experimental Start processing tests Configure project Each . represents 1024 bytes of output . Size of output: 0K Build project Each symbol represents 1024 bytes of output. '!' represents an error and '*' a warning. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090513/778d7fd8/attachment.html From hayes at bwh.harvard.edu Thu May 14 02:40:25 2009 From: hayes at bwh.harvard.edu (Kathryn Hayes) Date: Thu, 14 May 2009 02:40:25 -0400 (EDT) Subject: [slicer-devel] Compiling still... In-Reply-To: <81869fe90905132319i5258033dn4ca709c9d596cbc6@mail.gmail.com> References: <81869fe90905132020s1c3c3799y3e86ea78b4e33962@mail.gmail.com> <81869fe90905132102s84d8844ra85f1dfa826f990e@mail.gmail.com> <81869fe90905132132lc35f666u356a90783061499a@mail.gmail.com> <81869fe90905132222o25176b42s21961da89c44175c@mail.gmail.com> <81869fe90905132319i5258033dn4ca709c9d596cbc6@mail.gmail.com> Message-ID: On Windows, it's normal to not get a lot of feedback during the build process. It is probably still compiling, and will be done in a few hours. Does it look like it's using a fairly large chunk of CPU in Task Manager? Katie On Wed, 13 May 2009, Raquel Torres Peralta wrote: > Hi everyone. > It's been an hour since the getbuildtest.tcl stopped in the next message and > has not moved. > > Any suggestions? > > Best wishes, > Raquel > > > running: > D:/cygwin/home/ADMINI~1/Slicer/Slicer3/../Slicer3-lib/CMake-build/bin/ > test -D Experimental -C Debug > Site: CHEWI > Build name: Win32-vs8 > Create new tag: 20090514-0444 - Experimental > Start processing tests > Configure project > Each . represents 1024 bytes of output > . Size of output: 0K > Build project > Each symbol represents 1024 bytes of output. > '!' represents an error and '*' a warning. > From pieper at bwh.harvard.edu Thu May 14 07:43:57 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Thu, 14 May 2009 07:43:57 -0400 Subject: [slicer-devel] Python Surface ICP Registration In-Reply-To: References: Message-ID: <4A0C03FD.3060503@bwh.harvard.edu> Hi Harish - If you have a version of slicer that can run the python-based surface ICP module, then that version has the interpreter built in and you can write new modules for it (the recent slicer.org binaries should all be python-enabled). See this link for more info on using python: http://www.slicer.org/slicerWiki/index.php/Slicer3:Python If you are running into problems building slicer with python on your local machine, please send the full build log so we can track down what's going on. The python build process seems to get derailed by system specific issues we haven't pinned down yet. -Steve Harish Doddi wrote: > > > Hi All, > > We are looking into the python icp registration module developed by > Luca Antiga and Daniel Blezek and tried to register two femur models of > two patients. The results with affine registration look promising. We > would like to extend the functionality of this module to include MR/CT > images as target surface. We have an initial idea of how we should go > about doing this. > > I see the code SurfaceICPRegistration.py in the directory > Slicer3/Modules/Python. But I know that Slicer has a bug when building > with python modules. So I have been turning the python option off while > building slicer. As a result, my slicer executable doesnt have python > modules. Could someone please guide me in understanding how to > add/compile the code, integrating into slicer and also provide some > links where I can use python with ITK, VTK. > > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From norman-k-williams at uiowa.edu Thu May 14 11:59:57 2009 From: norman-k-williams at uiowa.edu (kent williams) Date: Thu, 14 May 2009 10:59:57 -0500 Subject: [slicer-devel] How are fiducial points displayed manipulated In-Reply-To: <9ddb27260905110651i1e3817e2he1916c3d8142a719@mail.gmail.com> Message-ID: Thanks all for responses. I am working on my own version of the ?one-model, many-interactors? pattern, which btw is kind of a challenge in vtk. I have a vtkHandleWidget in each view, and when any vtkHandleWidget is moved by the user, it updates the position of the other handle widgets in the other views. And when it does that, it decides whether or not they intersect the current slice plane in that view, and if it doesn?t it turns them off. Plus, when the slice is changed for the view, I traverse the point sets (and ultimately their points), and turn off the handlewidgets if they?re not in the new plane, and turn ON the handles if they ARE in plane. Plus, the Point Sets (and the Points within the Point Sets) are named, and I keep a tree view and menu bar menu up to date with the lists of names... Bottom line ? I have a lot of different observers doing all sorts of things, and a gazillion different objects all interacting. It?s a challenge to do, and even more of a challenge to structure and document it so that anyone else will be able to understand it after the fact ... On 5/11/09 8:51 AM, "Karthik Krishnan" wrote: > Kent: > > To answer your question specifically, as you've pointed out, VTK widgets can > only be associated with a single renderwindowinteractor, which in turn is > associated with a single render window. > > Slicer essentially uses a Model View design pattern. The MRML representation > is the model and the views (render windows) listen to events etc to > coordinate. In particular the hierarchy looks like? > > ? vtkMRMLScene > ???? | > ???? ---- vtkMRMLFiducialListNode > ?????????????????? | > ?????????????????? ------ fiducials > > > As Steve already pointed out, the 2D fiducials do not use the VTK widgets, > although I don't think there are significant deficiencies in the VTK widgets. > Highlighting as well as cursor change of nodes on mouse over can be quite > easily built into the VTK widgets. Several widgets do this, for instance, > vtkParallelopipedWidget, vtkContourWidget etc. The 3D ones use the old > vtkPointWidget. > > In the end, you will probably need to write your own model, much like MRML, to > listen to the VTK widgets for interaction events and update all the other > widgets that represent the same model in different views (render windows) > yourself. You may reuse the MRML library, if its generic enough to suite your > needs. > > > ? > > > > On Mon, May 11, 2009 at 9:36 AM, Steve Pieper wrote: >> Hi Kent - >> >> The short answer is you can have a look at Base/GUI/Tcl/SeedSWidget.tcl >> and FiducialsSWidget.tcl. >> >> Longer answer is that for the 2D slice viewers in slicer, the >> customization of VTK for slicer's purposes is implemented in incrTcl >> (object oriented tcl) with a base class called SWidget (Slicer Widget). >> ?These classes make the link between vtk-based primitives (events, >> actors, etc) and slicer's application structures like vtkSlicerSliceGUI, >> vtkMRMLFiducialList, etc. ?The main reasons for this were developer >> preference (mine!) for being able to reload the tcl scripts while the >> program is running in order to redefine the behavior of the widgets and >> because the vtk widgets didn't have the basic behavior I needed (like >> they didn't highlight on mouseover, which always bothered me). ?SWidget >> subclasses are also heavily used in the Editor module. >> >> In the 3D viewer, we do use vtk widgets, for example, >> Base/GUI/vtkSlicerFidudicalListWidget uses vtkPointWidgets for 3D >> interaction. ?Also, Base/GUI/Tcl/SlicePlaneSWidget.tcl manages a 3D vtk >> widget for interactive slice plane manipulation. >> >> In the end, since slicer internally considers the MRML representation as >> the 'definitive owner' of the fiducials state, there can be many >> different graphical editors (SWidgets in 2D, PointWidgets in 3D, and >> KWWidgets in the Fiducials Module). ?As long as they all track events >> correctly they cooperate. >> >> HTH, >> -Steve >> >> >> >> kent williams wrote: >>> > Hey. For our non-slicer tracing program, I've been banging my head on >>> using >>> > the vtkSeedWidget in order to allow points to be chosen for the purposes >>> of >>> > registration, and other diverse purposes. >>> > >>> > Hans Johnson (my boss/colleague) said "can you manipulate the point in all 3 >>> > views?" (i.e. axial sagittal, coronal), and I had to say 'No, the Seed >>> > Widget is something associated with a single renderer. ?The points >>> displayed >>> > on the off-plane views are actors without any corresponding interactor.' >>> > >>> > Hans Johnson just said 'well Slicer has fiducial points that you can move >>> in >>> > any of the 3 planer views -- you should look at how Slicer does it.' >>> > >>> > So I just spent a half hour wandering around in the Slicer source without >>> > really seeing how Slicer does it. ?It obviously does not use the >>> > vtkSeedWidget! >>> > >>> > Anyone have a clue as to where to look to find the slicer Fiducial point >>> > stuff? >>> > >>> > _______________________________________________ >>> > slicer-devel mailing list >>> > slicer-devel at bwh.harvard.edu >>> > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> > To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090514/e1bebe3b/attachment.html From nrannou at bwh.harvard.edu Thu May 14 14:27:21 2009 From: nrannou at bwh.harvard.edu (Nicolas Rannou) Date: Thu, 14 May 2009 14:27:21 -0400 Subject: [slicer-devel] Vizualisation of a slice extracted with vtkImageClip Message-ID: <4A0C6289.5040309@bwh.harvard.edu> Hi all, I'm trying to visualize a slice I extracted from the input volume without success. I extract the slice using vtkImageClip, then copy the slice into a vtkImageData (in a first method of my logic.cxx file): /---------------------------------------------------------------------------/ double bounds[6]; inVolume->GetImageData()->GetBounds(bounds); this->ImageClip = vtkImageClip::New(); this->ImageClip->SetInput(inVolume->GetImageData()); this->ImageClip->ClipDataOn(); // (same result if OFF) this->ImageClip->SetOutputWholeExtent(bounds[1]/2, bounds[1]/2,bounds[2], bounds[3],bounds[4], bounds[5]); this->ImageClip->Update(); this->CurrentSlide = vtkImageData::New(); this->CurrentSlide->DeepCopy( this->ImageClip->GetOutput()); /---------------------------------------------------------------------------/ (dimensionality of the output data (CurrentSlide) = 2 as expected, the size is good too) but when I want to see the extracted slice, I can't visualize the slice properly (cf attached screen shot). The content seems good, when I move on the red and green slides, but there should be the slice I extracted in the yellow slide. I set the extracted slide as output volume as below (in another method of the logic.cxx file): /---------------------------------------------------------------------------/ // copy RASToIJK matrix, and other attributes from input to output std::string name (outVolume->GetName()); std::string id (outVolume->GetID()); outVolume->CopyOrientation(inVolume); outVolume->SetAndObserveTransformNodeID(inVolume->GetTransformNodeID()); outVolume->SetName(name.c_str()); outVolume->SetAndObserveImageData( this->CurrentSlide); outVolume->SetModifiedSinceRead(1); /---------------------------------------------------------------------------/ I tried to extract the slice with vtkExtractVOI and vtkImageReslice too, but the result is the same. I must be missing something but I can't find out what. Thanks, Nicolas -------------- next part -------------- A non-text attachment was scrubbed... Name: slice_vizualization.png Type: image/png Size: 53995 bytes Desc: not available Url : http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090514/ecd93092/attachment-0001.png From tokuda at bwh.harvard.edu Thu May 14 14:51:05 2009 From: tokuda at bwh.harvard.edu (Junichi Tokuda) Date: Thu, 14 May 2009 14:51:05 -0400 Subject: [slicer-devel] Real-time frame grabbing in Slicer In-Reply-To: <4A0B8DF1.6040109@bwh.harvard.edu> References: <4A0B8DF1.6040109@bwh.harvard.edu> Message-ID: Hi Danielle, I think there is no frame grabbing function available in 3D Slicer at the moment. Of course the OpenIGTLink module may be used for this purpose, but I've never tried image transfer at 30 Hz. In the current version, the module imports images from the OpenIGTLink message buffer to the MRML every 100ms, thus frame rate higher than 10 Hz does not make sense. If your PC has sufficient performance, you could try reducing this interval by modifying the code at L.1159 in Modules/OpenIGTLinkIF/vtkOpenIGTLinkIFGUI.cxx. Another possibility is to use the next version of the neuro-endoscope module, which imports endoscope video images from the video grabber. Christoph Ruetz created the module for Slicer 2.6 and Sebastien from Kitware is now converting it for Slicer 3, according to Haiying. You may ask him about the current status. -Junichi On May 13, 2009, at 11:20 PM, Haiying Liu wrote: > Hi Danielle, > > Let's check with Junichi about real-time imaging support in Slicer. > > Junichi, does ProstateNav module in Slicer support frame grabbing? > > Best, > > Haiying > > Danielle Pace wrote: > >> Hi everyone, >> >> Our lab relies heavily on real-time imaging (for example, >> ultrasound and laparoscopy) for surgical guidance during various >> forms of interventions, which we have previously accomplished quite >> easily using Matrox frame grabbers and vtkMILVideoSource. I was >> wondering if Slicer also had a module that implements this >> functionality ? from my search online and through the Slicer source >> code, it doesn?t seem like it does. I understand that the >> OpenIGTLink protocol could be used to interface to a small IGSTK >> program using IGSTK?s frame grabbing functionality ? is this the >> recommended approach? I wonder if it?s overkill to have two >> programs running for something so simple as video grabbing, and >> whether this would introduce lag into the system. >> >> >> Thanks, >> >> Danielle >> >> -- >> >> Danielle Pace, BCmpH >> >> Masters Candidate, Biomedical Engineering >> >> Imaging Research Laboratories, Robarts Research Institute >> >> The University of Western Ontario, London, Ontario, Canada >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu >> with unsubscribe as the subject >> > From hliu at bwh.harvard.edu Thu May 14 16:31:22 2009 From: hliu at bwh.harvard.edu (Haiying Liu) Date: Thu, 14 May 2009 16:31:22 -0400 (EDT) Subject: [slicer-devel] Real-time frame grabbing in Slicer In-Reply-To: References: <4A0B8DF1.6040109@bwh.harvard.edu> Message-ID: On Thu, 14 May 2009, Junichi Tokuda wrote: > Hi Danielle, > > I think there is no frame grabbing function available in 3D Slicer at the > moment. > Of course the OpenIGTLink module may be used for this purpose, > but I've never tried image transfer at 30 Hz. In the current version, > the module imports images from the OpenIGTLink message buffer to > the MRML every 100ms, thus frame rate higher than 10 Hz does not > make sense. If your PC has sufficient performance, you could try reducing > this interval by modifying the code at L.1159 > in Modules/OpenIGTLinkIF/vtkOpenIGTLinkIFGUI.cxx. > > Another possibility is to use the next version of the neuro-endoscope module, > which imports endoscope video images from the video grabber. > Christoph Ruetz created the module for Slicer 2.6 and Sebastien from Kitware > is now converting it for Slicer 3, according to Haiying. > You may ask him about the current status. > We don't know yet what the neuro endoscope module looks like in Slicer3. We will find out at MIT program week in June. > -Junichi > > > On May 13, 2009, at 11:20 PM, Haiying Liu wrote: > >> Hi Danielle, >> >> Let's check with Junichi about real-time imaging support in Slicer. >> >> Junichi, does ProstateNav module in Slicer support frame grabbing? >> >> Best, >> >> Haiying >> >> Danielle Pace wrote: >> >>> Hi everyone, >>> >>> Our lab relies heavily on real-time imaging (for example, ultrasound and >>> laparoscopy) for surgical guidance during various forms of interventions, >>> which we have previously accomplished quite easily using Matrox frame >>> grabbers and vtkMILVideoSource. I was wondering if Slicer also had a >>> module that implements this functionality ? from my search online and >>> through the Slicer source code, it doesn?t seem like it does. I understand >>> that the OpenIGTLink protocol could be used to interface to a small IGSTK >>> program using IGSTK?s frame grabbing functionality ? is this the >>> recommended approach? I wonder if it?s overkill to have two programs >>> running for something so simple as video grabbing, and whether this would >>> introduce lag into the system. >>> >>> >>> Thanks, >>> >>> Danielle >>> >>> -- >>> >>> Danielle Pace, BCmpH >>> >>> Masters Candidate, Biomedical Engineering >>> >>> Imaging Research Laboratories, Robarts Research Institute >>> >>> The University of Western Ontario, London, Ontario, Canada >>> >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to >>> slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the >>> subject >>> >> > From harish.slicer at gmail.com Thu May 14 20:40:47 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Thu, 14 May 2009 17:40:47 -0700 Subject: [slicer-devel] python numpy installation Message-ID: Hi all, I have upgraded my python to python3.0 version and I am trying to install numpy on my linux machine. I have downloaded the package numpy from numpy.scipy.org and tried to run "python3.0 setup.py build --fcompiler=gnu" but I get the following error File "setup.py", line 62 print " --- Could not run svn info --- " ^ SyntaxError: invalid syntax Is this the correct way to install numpy module in python? Thanks, Harish -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090514/55ee40bc/attachment.html From pieper at bwh.harvard.edu Fri May 15 09:46:54 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 15 May 2009 09:46:54 -0400 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: References: Message-ID: <4A0D724E.5000903@bwh.harvard.edu> Hi Harish - Binary downloads of slicer come with numpy pre-compiled (bring up the python console and type 'import numpy'). If you build from source numpy is also enabled by default, although there are still issues building on some platforms. (If you get build errors, generate a build log following the instructions here and send the results to the slicer-devel list http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions#Errors_from_getbuildtest) For other python packages, I have had luck downloading binary packages from scipy.org and loading them into slicer. On the mac you can do something like this: > > import sys > d = '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] > for p in packages: > sys.path.append(d+p) To build other python packages, they can typically be built with this kind of line (at least on mac/linux): > > ../Slicer3 --launch ../../Slicer3-lib/python-build/bin/python setup.py install Best, Steve Harish Doddi wrote: > > > Hi all, > > I have upgraded my python to python3.0 version and I am trying to > install numpy on my linux machine. I have downloaded the package numpy > from numpy.scipy.org and tried to run > > "python3.0 setup.py build --fcompiler=gnu" > > but I get the following error > > File "setup.py", line 62 > print " --- Could not run svn info --- " > ^ > SyntaxError: invalid syntax > > > Is this the correct way to install numpy module in python? > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject From tokuda at bwh.harvard.edu Fri May 15 10:25:09 2009 From: tokuda at bwh.harvard.edu (Junichi Tokuda) Date: Fri, 15 May 2009 10:25:09 -0400 Subject: [slicer-devel] Node selector bug in 3.4RC and 3.5 Message-ID: Hi, I found a bug in the node selector widget in Slicer 3.4RC and 3.5 for Mac and Linux (I've not tested on windows). When I loaded a volume on Slicer and went to the Editor module, "Source Volume" menu kept resizing and I couldn't interact with the GUI anymore. My colleagues found the reproducibility depends on the font size in the application setting and the length of node name. (in case of my Mac OS X 10.5, it happened when I set the font size large) We observed the same bug on ver. 3.5 for Linux 64bit and ver. 3.4RC and 3.5 for Intel Darwin. -Junichi P.S. Drs. Oguro and Kekhia from BWH helped me reproduce the problem. Thanks! From jvs at bwh.harvard.edu Fri May 15 10:30:40 2009 From: jvs at bwh.harvard.edu (Julien von Siebenthal) Date: Fri, 15 May 2009 10:30:40 -0400 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: <4A0D724E.5000903@bwh.harvard.edu> References: <4A0D724E.5000903@bwh.harvard.edu> Message-ID: <4A0D7C90.4060509@bwh.harvard.edu> Hey, Just to say Python 3.0 may not work properly with numpy. You should check on the numpy site the latest supported version: I think it is python 2.6. By the way 2.6 must work perfectly, even 2.5. Best, Julien Steve Pieper wrote: > Hi Harish - > > Binary downloads of slicer come with numpy pre-compiled (bring up the > python console and type 'import numpy'). > > If you build from source numpy is also enabled by default, although > there are still issues building on some platforms. (If you get build > errors, generate a build log following the instructions here and send > the results to the slicer-devel list > http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions#Errors_from_getbuildtest) > > For other python packages, I have had luck downloading binary packages > from scipy.org and loading them into slicer. On the mac you can do > something like this: > > >> import sys >> d = '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' >> packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] >> for p in packages: >> sys.path.append(d+p) >> > > > To build other python packages, they can typically be built with this > kind of line (at least on mac/linux): > > >> ../Slicer3 --launch ../../Slicer3-lib/python-build/bin/python setup.py install >> > > Best, > Steve > > Harish Doddi wrote: > >> Hi all, >> >> I have upgraded my python to python3.0 version and I am trying to >> install numpy on my linux machine. I have downloaded the package numpy >> from numpy.scipy.org and tried to run >> >> "python3.0 setup.py build --fcompiler=gnu" >> >> but I get the following error >> >> File "setup.py", line 62 >> print " --- Could not run svn info --- " >> ^ >> SyntaxError: invalid syntax >> >> >> Is this the correct way to install numpy module in python? >> >> Thanks, >> Harish >> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> slicer-users mailing list >> slicer-users at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >> To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject >> > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to slicer-users-request at massmail.spl.harvard.edu with unsubscribe as the subject > -- Julien von Siebenthal, Ph.D. Psychiatric Neuroimaging Laboratory, Harvard Medical School Brigham and Women's Hospital, 75 Francis St., Boston MA 02115 jvs at bwh.harvard.edu From pieper at bwh.harvard.edu Fri May 15 10:41:35 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 15 May 2009 10:41:35 -0400 Subject: [slicer-devel] Node selector bug in 3.4RC and 3.5 In-Reply-To: References: Message-ID: <4A0D7F1F.5010301@bwh.harvard.edu> Hi Junichi - Ron and I started seeing the same thing - I have some ideas about what's going on (possibly an interaction with the kwwidget balloon help tooltips). I have a fix in mind and if it works I'll check it in later today... -Steve Junichi Tokuda wrote: > Hi, > > I found a bug in the node selector widget in Slicer 3.4RC and 3.5 > for Mac and Linux (I've not tested on windows). > > When I loaded a volume on Slicer and went to the Editor module, > "Source Volume" menu kept resizing and I couldn't interact with > the GUI anymore. > My colleagues found the reproducibility depends on > the font size in the application setting and the length of node name. > (in case of my Mac OS X 10.5, it happened when I set the font size > large) > We observed the same bug on ver. 3.5 for Linux 64bit and > ver. 3.4RC and 3.5 for Intel Darwin. > > -Junichi > > P.S. Drs. Oguro and Kekhia from BWH helped me reproduce the problem. > Thanks! > > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject From demian.wassermann at sophia.inria.fr Fri May 15 11:26:43 2009 From: demian.wassermann at sophia.inria.fr (Demian Wassermann) Date: Fri, 15 May 2009 17:26:43 +0200 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: <4A0D724E.5000903@bwh.harvard.edu> References: <4A0D724E.5000903@bwh.harvard.edu> Message-ID: <033D3F89-D47E-4672-93BE-C4E4DC673ED7@sophia.inria.fr> Hi Harish and steve, two things First, Harish you're trying to use python3.0 which is not yet a stable python version and has a lot of language changes, most of the numerical packages have not been put up to date for this. Slicer uses python 2.5, the most widely used stable version now. Steve, I think that if you do import sys d = '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/ site-packages/' sys.path.append( d ) then doing import matplotlib should work, if not a more pythonic version of your code could be import sys d = '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/ site-packages/' packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] sys.path += [ d+p for p in packages ] Cheers Dem -- Demian Wassermann demian.wassermann at sophia.inria.fr PhD Student Odyssee Research Project INRIA Sophia-Antipolis 2004 route des lucioles - FR-06902 On May 15, 2009, at 3:46 PM, Steve Pieper wrote: > Hi Harish - > > Binary downloads of slicer come with numpy pre-compiled (bring up the > python console and type 'import numpy'). > > If you build from source numpy is also enabled by default, although > there are still issues building on some platforms. (If you get build > errors, generate a build log following the instructions here and send > the results to the slicer-devel list > http://www.slicer.org/slicerWiki/index.php/ > Slicer3:Build_Instructions#Errors_from_getbuildtest) > > For other python packages, I have had luck downloading binary packages > from scipy.org and loading them into slicer. On the mac you can do > something like this: > >> >> import sys >> d = '/Library/Frameworks/Python.framework/Versions/2.5/lib/ >> python2.5/site-packages/' >> packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] >> for p in packages: >> sys.path.append(d+p) > > > To build other python packages, they can typically be built with this > kind of line (at least on mac/linux): > >> >> ../Slicer3 --launch ../../Slicer3-lib/python-build/bin/python >> setup.py install > > Best, > Steve > > Harish Doddi wrote: >> >> >> Hi all, >> >> I have upgraded my python to python3.0 version and I am trying to >> install numpy on my linux machine. I have downloaded the package >> numpy >> from numpy.scipy.org and tried to run >> >> "python3.0 setup.py build --fcompiler=gnu" >> >> but I get the following error >> >> File "setup.py", line 62 >> print " --- Could not run svn info --- " >> ^ >> SyntaxError: invalid syntax >> >> >> Is this the correct way to install numpy module in python? >> >> Thanks, >> Harish >> >> >> --------------------------------------------------------------------- >> --- >> >> _______________________________________________ >> slicer-users mailing list >> slicer-users at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >> To unsubscribe: send email to slicer-users- >> request at massmail.spl.harvard.edu with unsubscribe as the subject > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel- > request at massmail.spl.harvard.edu with unsubscribe as the subject From harish.slicer at gmail.com Fri May 15 13:55:41 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Fri, 15 May 2009 10:55:41 -0700 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: <033D3F89-D47E-4672-93BE-C4E4DC673ED7@sophia.inria.fr> References: <4A0D724E.5000903@bwh.harvard.edu> <033D3F89-D47E-4672-93BE-C4E4DC673ED7@sophia.inria.fr> Message-ID: Hi all, Thanks for the info. I am now trying to install python2.5 and also I need to install scipy plugin in it. Then I will follow the steps specified to link Python in slicer with scipy module. Thanks, Harish On Fri, May 15, 2009 at 8:26 AM, Demian Wassermann < demian.wassermann at sophia.inria.fr> wrote: > Hi Harish and steve, > > > two things > First, Harish you're trying to use python3.0 which is not yet a stable > python version and has a lot of language changes, most of the numerical > packages have not been put up to date for this. > > Slicer uses python 2.5, the most widely used stable version now. > > Steve, I think that if you do > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > sys.path.append( d ) > > then doing > > import matplotlib > > should work, > > if not a more pythonic version of your code could be > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] > sys.path += [ d+p for p in packages ] > > Cheers > Dem > > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > > > > > On May 15, 2009, at 3:46 PM, Steve Pieper wrote: > > Hi Harish - >> >> Binary downloads of slicer come with numpy pre-compiled (bring up the >> python console and type 'import numpy'). >> >> If you build from source numpy is also enabled by default, although >> there are still issues building on some platforms. (If you get build >> errors, generate a build log following the instructions here and send >> the results to the slicer-devel list >> http://www.slicer.org/slicerWiki/index.php/ >> Slicer3:Build_Instructions#Errors_from_getbuildtest) >> >> For other python packages, I have had luck downloading binary packages >> from scipy.org and loading them into slicer. On the mac you can do >> something like this: >> >> >>> import sys >>> d = >>> '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' >>> packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] >>> for p in packages: >>> sys.path.append(d+p) >>> >> >> >> To build other python packages, they can typically be built with this >> kind of line (at least on mac/linux): >> >> >>> ../Slicer3 --launch ../../Slicer3-lib/python-build/bin/python setup.py >>> install >>> >> >> Best, >> Steve >> >> Harish Doddi wrote: >> >>> >>> >>> Hi all, >>> >>> I have upgraded my python to python3.0 version and I am trying to >>> install numpy on my linux machine. I have downloaded the package numpy >>> from numpy.scipy.org and tried to run >>> >>> "python3.0 setup.py build --fcompiler=gnu" >>> >>> but I get the following error >>> >>> File "setup.py", line 62 >>> print " --- Could not run svn info --- " >>> ^ >>> SyntaxError: invalid syntax >>> >>> >>> Is this the correct way to install numpy module in python? >>> >>> Thanks, >>> Harish >>> >>> >>> ------------------------------------------------------------------------ >>> >>> _______________________________________________ >>> slicer-users mailing list >>> slicer-users at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >>> To unsubscribe: send email to slicer-users- >>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >> _______________________________________________ >> slicer-devel mailing list >> slicer-devel at bwh.harvard.edu >> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >> To unsubscribe: send email to slicer-devel- >> request at massmail.spl.harvard.edu with unsubscribe as the subject >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090515/8f30a0f2/attachment.html From harish.slicer at gmail.com Fri May 15 15:25:08 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Fri, 15 May 2009 12:25:08 -0700 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: References: <4A0D724E.5000903@bwh.harvard.edu> <033D3F89-D47E-4672-93BE-C4E4DC673ED7@sophia.inria.fr> Message-ID: Hi all, I tried installing scipy module into slicer python. I downloaded scipy setup.py file and used the folowing command to install into slicer. ./Slicer3 --launch ./lib/Python/bin/python /home/harishd/Python_Numpy/scipy-0.7.0/setup.py install Please note that I didnt use the slicer source code but just downloaded the executable from slicer.org But when I run the above command, there is a bunch of output produced along with the following information In file included from /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/Python.h:8, from scipy/cluster/src/vq.h:4, from scipy/cluster/src/vq.c:11: /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/pyconfig.h:939:1: warning: "_POSIX_C_SOURCE" redefined In file included from /usr/include/math.h:27, from scipy/cluster/src/vq.c:9: /usr/include/features.h:190:1: warning: this is the location of the previous definition /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/lib/python2.5/site-packages/numpy/core/include/numpy/__multiarray_api.h:959: warning: '_import_array' defined but not used gcc: scipy/cluster/src/vq_module.c gcc -pthread -shared build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so /usr/bin/ld: cannot find -lpython2.5 collect2: ld returned 1 exit status /usr/bin/ld: cannot find -lpython2.5 collect2: ld returned 1 exit status error: Command "gcc -pthread -shared build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so" failed with exit status 1 child process exited abnormally So Slicer is not able to start now. It complains me of the above error whenever it is lanuched. Can anybody help me in figuring out what the problem is? Thanks, Harish On Fri, May 15, 2009 at 10:55 AM, Harish Doddi wrote: > Hi all, > > Thanks for the info. I am now trying to install python2.5 and also I need > to install scipy plugin in it. Then I will follow the steps specified to > link Python in slicer with scipy module. > > > Thanks, > Harish > > > On Fri, May 15, 2009 at 8:26 AM, Demian Wassermann < > demian.wassermann at sophia.inria.fr> wrote: > >> Hi Harish and steve, >> >> >> two things >> First, Harish you're trying to use python3.0 which is not yet a stable >> python version and has a lot of language changes, most of the numerical >> packages have not been put up to date for this. >> >> Slicer uses python 2.5, the most widely used stable version now. >> >> Steve, I think that if you do >> >> import sys >> d = >> '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' >> sys.path.append( d ) >> >> then doing >> >> import matplotlib >> >> should work, >> >> if not a more pythonic version of your code could be >> >> import sys >> d = >> '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' >> packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] >> sys.path += [ d+p for p in packages ] >> >> Cheers >> Dem >> >> >> -- >> Demian Wassermann >> demian.wassermann at sophia.inria.fr >> PhD Student >> Odyssee Research Project >> INRIA Sophia-Antipolis >> 2004 route des lucioles - FR-06902 >> >> >> >> >> >> On May 15, 2009, at 3:46 PM, Steve Pieper wrote: >> >> Hi Harish - >>> >>> Binary downloads of slicer come with numpy pre-compiled (bring up the >>> python console and type 'import numpy'). >>> >>> If you build from source numpy is also enabled by default, although >>> there are still issues building on some platforms. (If you get build >>> errors, generate a build log following the instructions here and send >>> the results to the slicer-devel list >>> http://www.slicer.org/slicerWiki/index.php/ >>> Slicer3:Build_Instructions#Errors_from_getbuildtest) >>> >>> For other python packages, I have had luck downloading binary packages >>> from scipy.org and loading them into slicer. On the mac you can do >>> something like this: >>> >>> >>>> import sys >>>> d = >>>> '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' >>>> packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] >>>> for p in packages: >>>> sys.path.append(d+p) >>>> >>> >>> >>> To build other python packages, they can typically be built with this >>> kind of line (at least on mac/linux): >>> >>> >>>> ../Slicer3 --launch ../../Slicer3-lib/python-build/bin/python setup.py >>>> install >>>> >>> >>> Best, >>> Steve >>> >>> Harish Doddi wrote: >>> >>>> >>>> >>>> Hi all, >>>> >>>> I have upgraded my python to python3.0 version and I am trying to >>>> install numpy on my linux machine. I have downloaded the package numpy >>>> from numpy.scipy.org and tried to run >>>> >>>> "python3.0 setup.py build --fcompiler=gnu" >>>> >>>> but I get the following error >>>> >>>> File "setup.py", line 62 >>>> print " --- Could not run svn info --- " >>>> ^ >>>> SyntaxError: invalid syntax >>>> >>>> >>>> Is this the correct way to install numpy module in python? >>>> >>>> Thanks, >>>> Harish >>>> >>>> >>>> ------------------------------------------------------------------------ >>>> >>>> _______________________________________________ >>>> slicer-users mailing list >>>> slicer-users at bwh.harvard.edu >>>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users >>>> To unsubscribe: send email to slicer-users- >>>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>>> >>> _______________________________________________ >>> slicer-devel mailing list >>> slicer-devel at bwh.harvard.edu >>> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel >>> To unsubscribe: send email to slicer-devel- >>> request at massmail.spl.harvard.edu with unsubscribe as the subject >>> >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/pipermail/slicer-devel/attachments/20090515/7c86a99c/attachment-0001.html From jvs at bwh.harvard.edu Fri May 15 16:05:23 2009 From: jvs at bwh.harvard.edu (Julien von Siebenthal) Date: Fri, 15 May 2009 16:05:23 -0400 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: References: <4A0D724E.5000903@bwh.harvard.edu> <033D3F89-D47E-4672-93BE-C4E4DC673ED7@sophia.inria.fr> Message-ID: <4A0DCB03.4050401@bwh.harvard.edu> Hi Harish, First thing to try: looking at your output, you should not forget to set the PATH & LD_LIBRARY_PATH to your current python bin (/bin) and lib (/lib). Perhaps, it should fix the problem. Second thing to try: it is still challenging to install scipy properly. If you're a Mac or Linux user, I advice you to compile the Atlas library that offers a lapack installation. Then, with this lapack, you can compile a clean scipy that will fit your environment. I can give you more guidance if you want to try. Best, Julien Harish Doddi wrote: > Hi all, > > I tried installing scipy module into slicer python. > > I downloaded scipy setup.py file and used the folowing command to > install into slicer. > > ./Slicer3 --launch ./lib/Python/bin/python > /home/harishd/Python_Numpy/scipy-0.7.0/setup.py install > > Please note that I didnt use the slicer source code but just > downloaded the executable from slicer.org > > But when I run the above command, there is a bunch of output produced > along with the following information > > In file included from > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/Python.h:8, > from scipy/cluster/src/vq.h:4, > from scipy/cluster/src/vq.c:11: > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/pyconfig.h:939:1: > warning: "_POSIX_C_SOURCE" redefined > In file included from /usr/include/math.h:27, > from scipy/cluster/src/vq.c:9: > /usr/include/features.h:190:1: warning: this is the location of the > previous definition > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/lib/python2.5/site-packages/numpy/core/include/numpy/__multiarray_api.h:959: > warning: '_import_array' defined but not used > gcc: scipy/cluster/src/vq_module.c > gcc -pthread -shared > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so > /usr/bin/ld: cannot find -lpython2.5 > collect2: ld returned 1 exit status > /usr/bin/ld: cannot find -lpython2.5 > collect2: ld returned 1 exit status > error: Command "gcc -pthread -shared > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so" failed with exit status 1 > > child process exited abnormally > > So Slicer is not able to start now. It complains me of the above error > whenever it is lanuched. > > Can anybody help me in figuring out what the problem is? > > > Thanks, > Harish > > > > > > > On Fri, May 15, 2009 at 10:55 AM, Harish Doddi > > wrote: > > Hi all, > > Thanks for the info. I am now trying to install python2.5 and also > I need to install scipy plugin in it. Then I will follow the steps > specified to link Python in slicer with scipy module. > > > Thanks, > Harish > > > On Fri, May 15, 2009 at 8:26 AM, Demian Wassermann > > wrote: > > Hi Harish and steve, > > > two things > First, Harish you're trying to use python3.0 which is not yet > a stable python version and has a lot of language changes, > most of the numerical packages have not been put up to date > for this. > > Slicer uses python 2.5, the most widely used stable version now. > > Steve, I think that if you do > > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > sys.path.append( d ) > > then doing > > import matplotlib > > should work, > > if not a more pythonic version of your code could be > > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] > sys.path += [ d+p for p in packages ] > > Cheers > Dem > > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > > > > > On May 15, 2009, at 3:46 PM, Steve Pieper wrote: > > Hi Harish - > > Binary downloads of slicer come with numpy pre-compiled > (bring up the > python console and type 'import numpy'). > > If you build from source numpy is also enabled by default, > although > there are still issues building on some platforms. (If > you get build > errors, generate a build log following the instructions > here and send > the results to the slicer-devel list > http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions#Errors_from_getbuildtest) > > For other python packages, I have had luck downloading > binary packages > from scipy.org and loading them into > slicer. On the mac you can do > something like this: > > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > packages = ['matplotlib', 'neuroimaging', 'numpy', '', > 'scipy'] > for p in packages: > sys.path.append(d+p) > > > > To build other python packages, they can typically be > built with this > kind of line (at least on mac/linux): > > > ../Slicer3 --launch > ../../Slicer3-lib/python-build/bin/python setup.py install > > > Best, > Steve > > Harish Doddi wrote: > > > > Hi all, > > I have upgraded my python to python3.0 version and I > am trying to > install numpy on my linux machine. I have downloaded > the package numpy > from numpy.scipy.org > and tried to run > > "python3.0 setup.py build --fcompiler=gnu" > > but I get the following error > > File "setup.py", line 62 > print " --- Could not run svn info --- " > ^ > SyntaxError: invalid syntax > > > Is this the correct way to install numpy module in python? > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to > slicer-users-request at massmail.spl.harvard.edu > with > unsubscribe as the subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe > as the subject > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to slicer-devel-request at massmail.spl.harvard.edu with unsubscribe as the subject -- Julien von Siebenthal, Ph.D. Psychiatric Neuroimaging Laboratory, Harvard Medical School Brigham and Women's Hospital, 75 Francis St., Boston MA 02115 jvs at bwh.harvard.edu From pieper at bwh.harvard.edu Fri May 15 17:13:08 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Fri, 15 May 2009 17:13:08 -0400 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: References: <4A0D724E.5000903@bwh.harvard.edu> <033D3F89-D47E-4672-93BE-C4E4DC673ED7@sophia.inria.fr> Message-ID: <4A0DDAE4.8090008@bwh.harvard.edu> Hi Harish - I would suggest trying the binaries of scipy and adding them to the python path. Scipy is a pain to build. -Steve Harish Doddi wrote: > Hi all, > > I tried installing scipy module into slicer python. > > I downloaded scipy setup.py file and used the folowing command to > install into slicer. > > ./Slicer3 --launch ./lib/Python/bin/python > /home/harishd/Python_Numpy/scipy-0.7.0/setup.py install > > Please note that I didnt use the slicer source code but just downloaded > the executable from slicer.org > > But when I run the above command, there is a bunch of output produced > along with the following information > > In file included from > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/Python.h:8, > from scipy/cluster/src/vq.h:4, > from scipy/cluster/src/vq.c:11: > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/pyconfig.h:939:1: > warning: "_POSIX_C_SOURCE" redefined > In file included from /usr/include/math.h:27, > from scipy/cluster/src/vq.c:9: > /usr/include/features.h:190:1: warning: this is the location of the > previous definition > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/lib/python2.5/site-packages/numpy/core/include/numpy/__multiarray_api.h:959: > warning: '_import_array' defined but not used > gcc: scipy/cluster/src/vq_module.c > gcc -pthread -shared > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so > /usr/bin/ld: cannot find -lpython2.5 > collect2: ld returned 1 exit status > /usr/bin/ld: cannot find -lpython2.5 > collect2: ld returned 1 exit status > error: Command "gcc -pthread -shared > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so" failed with exit status 1 > > child process exited abnormally > > So Slicer is not able to start now. It complains me of the above error > whenever it is lanuched. > > Can anybody help me in figuring out what the problem is? > > > Thanks, > Harish > > > > > > > On Fri, May 15, 2009 at 10:55 AM, Harish Doddi > wrote: > > Hi all, > > Thanks for the info. I am now trying to install python2.5 and also I > need to install scipy plugin in it. Then I will follow the steps > specified to link Python in slicer with scipy module. > > > Thanks, > Harish > > > On Fri, May 15, 2009 at 8:26 AM, Demian Wassermann > > wrote: > > Hi Harish and steve, > > > two things > First, Harish you're trying to use python3.0 which is not yet a > stable python version and has a lot of language changes, most of > the numerical packages have not been put up to date for this. > > Slicer uses python 2.5, the most widely used stable version now. > > Steve, I think that if you do > > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > sys.path.append( d ) > > then doing > > import matplotlib > > should work, > > if not a more pythonic version of your code could be > > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > packages = ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] > sys.path += [ d+p for p in packages ] > > Cheers > Dem > > > -- > Demian Wassermann > demian.wassermann at sophia.inria.fr > > PhD Student > Odyssee Research Project > INRIA Sophia-Antipolis > 2004 route des lucioles - FR-06902 > > > > > > On May 15, 2009, at 3:46 PM, Steve Pieper wrote: > > Hi Harish - > > Binary downloads of slicer come with numpy pre-compiled > (bring up the > python console and type 'import numpy'). > > If you build from source numpy is also enabled by default, > although > there are still issues building on some platforms. (If you > get build > errors, generate a build log following the instructions here > and send > the results to the slicer-devel list > http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions#Errors_from_getbuildtest) > > For other python packages, I have had luck downloading > binary packages > from scipy.org and loading them into > slicer. On the mac you can do > something like this: > > > import sys > d = > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > packages = ['matplotlib', 'neuroimaging', 'numpy', '', > 'scipy'] > for p in packages: > sys.path.append(d+p) > > > > To build other python packages, they can typically be built > with this > kind of line (at least on mac/linux): > > > ../Slicer3 --launch > ../../Slicer3-lib/python-build/bin/python setup.py install > > > Best, > Steve > > Harish Doddi wrote: > > > > Hi all, > > I have upgraded my python to python3.0 version and I am > trying to > install numpy on my linux machine. I have downloaded the > package numpy > from numpy.scipy.org > and tried to run > > "python3.0 setup.py build --fcompiler=gnu" > > but I get the following error > > File "setup.py", line 62 > print " --- Could not run svn info --- " > ^ > SyntaxError: invalid syntax > > > Is this the correct way to install numpy module in python? > > Thanks, > Harish > > > ------------------------------------------------------------------------ > > _______________________________________________ > slicer-users mailing list > slicer-users at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users > To unsubscribe: send email to > slicer-users-request at massmail.spl.harvard.edu > with > unsubscribe as the subject > > _______________________________________________ > slicer-devel mailing list > slicer-devel at bwh.harvard.edu > > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-devel > To unsubscribe: send email to > slicer-devel-request at massmail.spl.harvard.edu > with unsubscribe > as the subject > > > > From nagy.attila at yahoo.com Fri May 15 17:32:32 2009 From: nagy.attila at yahoo.com (Attila Nagy) Date: Fri, 15 May 2009 14:32:32 -0700 (PDT) Subject: [slicer-devel] [slicer-users] python numpy installation Message-ID: <851081.34769.qm@web37403.mail.mud.yahoo.com> Harish, make sure that you have libpython2.5.so in the linker path. Not libpython25.so, for example. I remember to have had similar issues some time ago; look at slicer_variables.tcl for the correct python lib filename - on Solaris python was built as libpython25.so, and the build system was looking for lpython2.5.so (or the other way around :)) > > /usr/bin/ld: cannot find -lpython2.5 Do you have "build/temp.linux-x86_64-2.5" ? Hope it helps, Attila --- On Fri, 5/15/09, Steve Pieper wrote: > From: Steve Pieper > Subject: Re: [slicer-devel] [slicer-users] python numpy installation > To: "Harish Doddi" > Cc: slicer-users at bwh.harvard.edu, slicer-devel at bwh.harvard.edu > Date: Friday, May 15, 2009, 11:13 PM > Hi Harish - > > I would suggest trying the binaries of scipy and adding > them to the > python path.? Scipy is a pain to build. > > -Steve > > Harish Doddi wrote: > > Hi all, > > > > I tried installing scipy module into slicer python. > > > > I downloaded scipy setup.py file and used the folowing > command to > > install into slicer. > > > > ./Slicer3 --launch ./lib/Python/bin/python > > /home/harishd/Python_Numpy/scipy-0.7.0/setup.py > install > > > > Please note that I didnt use the slicer source code > but just downloaded > > the executable from slicer.org > > > > But when I run the above command, there is a bunch of > output produced > > along with the following information > > > > In file included from > > > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/Python.h:8, > >? ? ? ? ? ? ? ? > ? from scipy/cluster/src/vq.h:4, > >? ? ? ? ? ? ? ? > ? from scipy/cluster/src/vq.c:11: > > > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/pyconfig.h:939:1: > > > warning: "_POSIX_C_SOURCE" redefined > > In file included from /usr/include/math.h:27, > >? ? ? ? ? ? ? ? > ? from scipy/cluster/src/vq.c:9: > > /usr/include/features.h:190:1: warning: this is the > location of the > > previous definition > > > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/lib/python2.5/site-packages/numpy/core/include/numpy/__multiarray_api.h:959: > > > warning: '_import_array' defined but not used > > gcc: scipy/cluster/src/vq_module.c > > gcc -pthread -shared > > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > > > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > > > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so > > /usr/bin/ld: cannot find -lpython2.5 > > collect2: ld returned 1 exit status > > /usr/bin/ld: cannot find -lpython2.5 > > collect2: ld returned 1 exit status > > error: Command "gcc -pthread -shared > > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > > > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > > > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so" > failed with exit status 1 > > > > child process exited abnormally > > > > So Slicer is not able to start now. It complains me of > the above error > > whenever it is lanuched. > > > > Can anybody help me in figuring out what the problem > is? > > > > > > Thanks, > > Harish > > > > > > > > > > > > > > On Fri, May 15, 2009 at 10:55 AM, Harish Doddi > > > > wrote: > > > >? ???Hi all, > > > >? ???Thanks for the info. I am now > trying to install python2.5 and also I > >? ???need to install scipy plugin > in it. Then I will follow the steps > >? ???specified to link Python in > slicer with scipy module. > > > > > >? ???Thanks, > >? ???Harish > > > > > >? ???On Fri, May 15, 2009 at 8:26 > AM, Demian Wassermann > >? ??? >? ???> > wrote: > > > >? ? ? ???Hi Harish and > steve, > > > > > >? ? ? ???two things > >? ? ? ???First, Harish > you're trying to use python3.0 which is not yet a > >? ? ? ???stable python > version and has a lot of language changes, most of > >? ? ? ???the numerical > packages have not been put up to date for this. > > > >? ? ? ???Slicer uses > python 2.5, the most widely used stable version now. > > > >? ? ? ???Steve, I think > that if you do > > > > > >? ? ? ???import sys > >? ? ? ???d = > >? ? ? > ???'/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > >? ? ? ???sys.path.append( > d ) > > > >? ? ? ???then doing > > > >? ? ? ???import > matplotlib > > > >? ? ? ???should work, > > > >? ? ? ???if not a more > pythonic version of your code could be > > > > > >? ? ? ???import sys > >? ? ? ???d = > >? ? ? > ???'/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > >? ? ? ???packages = > ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] > >? ? ? ???sys.path += [ > d+p for p in packages ] > > > >? ? ? ???Cheers > >? ? ? ???Dem > > > > > >? ? ? ???-- > >? ? ? ???Demian > Wassermann > >? ? ? ???demian.wassermann at sophia.inria.fr > >? ? ? ??? > >? ? ? ???PhD Student > >? ? ? ???Odyssee Research > Project > >? ? ? ???INRIA > Sophia-Antipolis > >? ? ? ???2004 route des > lucioles - FR-06902 > > > > > > > > > > > >? ? ? ???On May 15, 2009, > at 3:46 PM, Steve Pieper wrote: > > > >? ? ? ? ? ???Hi > Harish - > > > >? ? ? ? ? > ???Binary downloads of slicer come with numpy > pre-compiled > >? ? ? ? ? > ???(bring up the > >? ? ? ? ? > ???python console and type 'import numpy'). > > > >? ? ? ? ? ???If > you build from source numpy is also enabled by default, > >? ? ? ? ? > ???although > >? ? ? ? ? > ???there are still issues building on some > platforms.? (If you > >? ? ? ? ? > ???get build > >? ? ? ? ? > ???errors, generate a build log following the > instructions here > >? ? ? ? ? > ???and send > >? ? ? ? ? > ???the results to the slicer-devel list > >? ? ? ? ? ???http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions#Errors_from_getbuildtest) > > > >? ? ? ? ? > ???For other python packages, I have had luck > downloading > >? ? ? ? ? > ???binary packages > >? ? ? ? ? > ???from scipy.org and loading them into > >? ? ? ? ? > ???slicer.? On the mac you can do > >? ? ? ? ? > ???something like this: > > > > > >? ? ? ? ? ? ? > ???import sys > >? ? ? ? ? ? ? > ???d = > >? ? ? ? ? ? ? > ???'/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > >? ? ? ? ? ? ? > ???packages = ['matplotlib', 'neuroimaging', > 'numpy', '', > >? ? ? ? ? ? ? > ???'scipy'] > >? ? ? ? ? ? ? > ???for p in packages: > >? ? ? ? ? ? ? ? > ? sys.path.append(d+p) > > > > > > > >? ? ? ? ? ???To > build other python packages, they can typically be built > >? ? ? ? ? > ???with this > >? ? ? ? ? > ???kind of line (at least on mac/linux): > > > > > >? ? ? ? ? ? ? > ???../Slicer3 --launch > >? ? ? ? ? ? ? > ???../../Slicer3-lib/python-build/bin/python > setup.py install > > > > > >? ? ? ? ? > ???Best, > >? ? ? ? ? > ???Steve > > > >? ? ? ? ? > ???Harish Doddi wrote: > > > > > > > >? ? ? ? ? ? ? > ???Hi all, > > > >? ? ? ? ? ? ? > ???I have upgraded my python to python3.0 > version and I am > >? ? ? ? ? ? ? > ???trying to > >? ? ? ? ? ? ? > ???install numpy on my linux machine. I have > downloaded the > >? ? ? ? ? ? ? > ???package numpy > >? ? ? ? ? ? ? > ???from numpy.scipy.org > >? ? ? ? ? ? ? > ??? and tried to > run > > > >? ? ? ? ? ? ? > ???"python3.0 setup.py build > --fcompiler=gnu" > > > >? ? ? ? ? ? ? > ???but I get the following error > > > >? ? ? ? ? ? ? ? > ? File "setup.py", line 62 > >? ? ? ? ? ? ? ? > ? ? print " --- Could not run svn info --- " > >? ? ? ? ? ? ? ? > ? ? ? ? ? ? ? ? > ? ? ? ? ? ? ? ? > ? ? ? ? ???^ > >? ? ? ? ? ? ? > ???SyntaxError: invalid syntax > > > > > >? ? ? ? ? ? ? > ???Is this the correct way to install numpy > module in python? > > > >? ? ? ? ? ? ? > ???Thanks, > >? ? ? ? ? ? ? > ???Harish > From harish.slicer at gmail.com Tue May 19 13:31:40 2009 From: harish.slicer at gmail.com (Harish Doddi) Date: Tue, 19 May 2009 10:31:40 -0700 Subject: [slicer-devel] [slicer-users] python numpy installation In-Reply-To: <851081.34769.qm@web37403.mail.mud.yahoo.com> References: <851081.34769.qm@web37403.mail.mud.yahoo.com> Message-ID: Hi all, This problem got fixed. Thanks all for your help. Harish On Fri, May 15, 2009 at 2:32 PM, Attila Nagy wrote: > > Harish, > > make sure that you have libpython2.5.so in the linker path. Not > libpython25.so, for example. I remember to have had similar issues some time > ago; look at slicer_variables.tcl for the correct python lib filename - on > Solaris python was built as libpython25.so, and the build system was looking > for lpython2.5.so (or the other way around :)) > > > > /usr/bin/ld: cannot find -lpython2.5 > > > Do you have "build/temp.linux-x86_64-2.5" ? > > Hope it helps, > > Attila > > --- On Fri, 5/15/09, Steve Pieper wrote: > > > From: Steve Pieper > > Subject: Re: [slicer-devel] [slicer-users] python numpy installation > > To: "Harish Doddi" > > Cc: slicer-users at bwh.harvard.edu, slicer-devel at bwh.harvard.edu > > Date: Friday, May 15, 2009, 11:13 PM > > Hi Harish - > > > > I would suggest trying the binaries of scipy and adding > > them to the > > python path. Scipy is a pain to build. > > > > -Steve > > > > Harish Doddi wrote: > > > Hi all, > > > > > > I tried installing scipy module into slicer python. > > > > > > I downloaded scipy setup.py file and used the folowing > > command to > > > install into slicer. > > > > > > ./Slicer3 --launch ./lib/Python/bin/python > > > /home/harishd/Python_Numpy/scipy-0.7.0/setup.py > > install > > > > > > Please note that I didnt use the slicer source code > > but just downloaded > > > the executable from slicer.org > > > > > > But when I run the above command, there is a bunch of > > output produced > > > along with the following information > > > > > > In file included from > > > > > > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/Python.h:8, > > > > > from scipy/cluster/src/vq.h:4, > > > > > from scipy/cluster/src/vq.c:11: > > > > > > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/include/python2.5/pyconfig.h:939:1: > > > > > warning: "_POSIX_C_SOURCE" redefined > > > In file included from /usr/include/math.h:27, > > > > > from scipy/cluster/src/vq.c:9: > > > /usr/include/features.h:190:1: warning: this is the > > location of the > > > previous definition > > > > > > /home/harishd/RA/slicer_exeutable/Slicer3-3.5-alpha-2009-05-08-linux-x86_64/lib/Python/lib/python2.5/site-packages/numpy/core/include/numpy/__multiarray_api.h:959: > > > > > warning: '_import_array' defined but not used > > > gcc: scipy/cluster/src/vq_module.c > > > gcc -pthread -shared > > > > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > > > > > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > > > > > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > > > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so > > > /usr/bin/ld: cannot find -lpython2.5 > > > collect2: ld returned 1 exit status > > > /usr/bin/ld: cannot find -lpython2.5 > > > collect2: ld returned 1 exit status > > > error: Command "gcc -pthread -shared > > > > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq_module.o > > > build/temp.linux-x86_64-2.5/scipy/cluster/src/vq.o > > > > > -L/home/hayes/Slicer3-release-gbt/Slicer3/../Slicer3-lib/python-build/lib > > > > > -Lbuild/temp.linux-x86_64-2.5 -lpython2.5 -o > > > build/lib.linux-x86_64-2.5/scipy/cluster/_vq.so" > > failed with exit status 1 > > > > > > child process exited abnormally > > > > > > So Slicer is not able to start now. It complains me of > > the above error > > > whenever it is lanuched. > > > > > > Can anybody help me in figuring out what the problem > > is? > > > > > > > > > Thanks, > > > Harish > > > > > > > > > > > > > > > > > > > > > On Fri, May 15, 2009 at 10:55 AM, Harish Doddi < > harish.slicer at gmail.com > > > > > > > > wrote: > > > > > > Hi all, > > > > > > Thanks for the info. I am now > > trying to install python2.5 and also I > > > need to install scipy plugin > > in it. Then I will follow the steps > > > specified to link Python in > > slicer with scipy module. > > > > > > > > > Thanks, > > > Harish > > > > > > > > > On Fri, May 15, 2009 at 8:26 > > AM, Demian Wassermann > > > > > > > > wrote: > > > > > > Hi Harish and > > steve, > > > > > > > > > two things > > > First, Harish > > you're trying to use python3.0 which is not yet a > > > stable python > > version and has a lot of language changes, most of > > > the numerical > > packages have not been put up to date for this. > > > > > > Slicer uses > > python 2.5, the most widely used stable version now. > > > > > > Steve, I think > > that if you do > > > > > > > > > import sys > > > d = > > > > > > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > > > sys.path.append( > > d ) > > > > > > then doing > > > > > > import > > matplotlib > > > > > > should work, > > > > > > if not a more > > pythonic version of your code could be > > > > > > > > > import sys > > > d = > > > > > > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > > > packages = > > ['matplotlib', 'neuroimaging', 'numpy', '', 'scipy'] > > > sys.path += [ > > d+p for p in packages ] > > > > > > Cheers > > > Dem > > > > > > > > > -- > > > Demian > > Wassermann > > > demian.wassermann at sophia.inria.fr > > > > > > PhD Student > > > Odyssee Research > > Project > > > INRIA > > Sophia-Antipolis > > > 2004 route des > > lucioles - FR-06902 > > > > > > > > > > > > > > > > > > On May 15, 2009, > > at 3:46 PM, Steve Pieper wrote: > > > > > > Hi > > Harish - > > > > > > > > Binary downloads of slicer come with numpy > > pre-compiled > > > > > (bring up the > > > > > python console and type 'import numpy'). > > > > > > If > > you build from source numpy is also enabled by default, > > > > > although > > > > > there are still issues building on some > > platforms. (If you > > > > > get build > > > > > errors, generate a build log following the > > instructions here > > > > > and send > > > > > the results to the slicer-devel list > > > > http://www.slicer.org/slicerWiki/index.php/Slicer3:Build_Instructions#Errors_from_getbuildtest > ) > > > > > > > > For other python packages, I have had luck > > downloading > > > > > binary packages > > > > > from scipy.org and loading them into > > > > > slicer. On the mac you can do > > > > > something like this: > > > > > > > > > > > import sys > > > > > d = > > > > > > '/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/' > > > > > packages = ['matplotlib', 'neuroimaging', > > 'numpy', '', > > > > > 'scipy'] > > > > > for p in packages: > > > > > sys.path.append(d+p) > > > > > > > > > > > > To > > build other python packages, they can typically be built > > > > > with this > > > > > kind of line (at least on mac/linux): > > > > > > > > > > > ../Slicer3 --launch > > > > > ../../Slicer3-lib/python-build/bin/python > > setup.py install > > > > > > > > > > > Best, > > > > > Steve > > > > > > > > Harish Doddi wrote: > > > > > > > > > > > > > > Hi all, > > > > > > > > I have upgraded my python to python3.0 > > version and I am > > > > > trying to > > > > > install numpy on my linux machine. I have > > downloaded the > > > > > package numpy > > > > > from numpy.scipy.org > > > > > and tried to > > run > > > > > > > > "python3.0 setup.py build > > --fcompiler=gnu" > > > > > > > > but I get the following error > > > > > > > > File "setup.py", line 62 > > > > > print " --- Could not run svn info --- " > > > > > > > > > ^ > > > > > SyntaxError: invalid syntax > > > > > > > > > > > Is this the correct way to install numpy > > module in python? > > > > > > > > Thanks, > > > > > Harish > > > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://massmail.spl.harvard.edu/mailman/private/slicer-devel/attachments/20090519/1be0bb5b/attachment.html From sebastien.barre at kitware.com Tue May 19 15:47:29 2009 From: sebastien.barre at kitware.com (Sebastien BARRE) Date: Tue, 19 May 2009 15:47:29 -0400 Subject: [slicer-devel] Error linking ChangeTrackerTest Message-ID: <20090519194736.05DEB621E@public.kitware.com> Hi guys, I've had this problem "for ever", but so far I had commented the ChangeTracker module. Yet, I would like to get to the bottom of this. This happens on the HEAD, or in the Slicer-3-4 branch. On Vista + Visual Studio 2008, shared DLL, etc. Sounds familiar to anyone? Linking CXX executable ../../../bin/ChangeTrackerTest.exe ITKNrrdIO.lib(miscAir.obj) : error LNK2005: itk_airFree already defined in CommandIO.lib(CommandIO.dll) ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceDimension already defined in CommandIO.lib(CommandIO.dll) ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceSet already defined in CommandIO.lib(CommandIO.dll) ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceDimensionSet already defined in CommandIO.lib(CommandIO.dll) ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceOriginSet already defined in CommandIO.lib(CommandIO.dll) I'm not using getbuiltest. I have at least 6 different build of Slicer3, I can only allocate some many hundreds of Gigabytes to the project, so I'm sharing the same VTK/ITK/KWW as much as I can between them, as I've done for years. This linking error though... Thanks From pieper at bwh.harvard.edu Tue May 19 16:42:58 2009 From: pieper at bwh.harvard.edu (Steve Pieper) Date: Tue, 19 May 2009 16:42:58 -0400 Subject: [slicer-devel] Error linking ChangeTrackerTest In-Reply-To: <20090519194736.05DEB621E@public.kitware.com> References: <20090519194736.05DEB621E@public.kitware.com> Message-ID: <4A1319D2.6080606@bwh.harvard.edu> Hi Seb - Can't explain that - it doesn't occur on my builds (getbuildtest, of course). What's odd is that CommandIO is an itk IO factory plugin and there's no reason for anything else to link with it. I don't see where in the cmake code the ChangeTrackerTest program would told to link with it... -Steve Sebastien BARRE wrote: > Hi guys, > > I've had this problem "for ever", but so far I had commented the > ChangeTracker module. Yet, I would like to get to the bottom of this. > This happens on the HEAD, or in the Slicer-3-4 branch. On Vista + Visual > Studio 2008, shared DLL, etc. > > Sounds familiar to anyone? > > Linking CXX executable ../../../bin/ChangeTrackerTest.exe > ITKNrrdIO.lib(miscAir.obj) : error LNK2005: itk_airFree already defined > in CommandIO.lib(CommandIO.dll) > ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceDimension > already defined in CommandIO.lib(CommandIO.dll) > ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceSet already > defined in CommandIO.lib(CommandIO.dll) > ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceDimensionSet > already defined in CommandIO.lib(CommandIO.dll) > ITKNrrdIO.lib(simple.obj) : error LNK2005: itk_nrrdSpaceOriginSet > already defined in CommandIO.lib(CommandIO.dll) >