From spmlxf at yahoo.com Mon Jul 3 03:42:40 2006 From: spmlxf at yahoo.com (xingfeng lee) Date: Mon, 3 Jul 2006 00:42:40 -0700 (PDT) Subject: [MINC-users] transformation of GE to MINC Message-ID: <20060703074240.87156.qmail@web50107.mail.yahoo.com> hello all: I encountered a problem when I transformate GE to MINC. we acquired data with GE-signa 1.5 Tesla scanner, and wanted ultimate data with MINC format. for this sake. we tested 2 ways: 1. transformate GE data to DICOM, then to MINC. At the ADW4.0 workstation of GE, exported the data of GE format to ADW to finish DICOM transformation. but there is a trouble that one folder can only hold 1000 files in GE equipment, 3000 files of one run in our experiment were divided into 3 folders, they can't be accepted by the programme of dcm2mnc(BIC MNI McGill). we tried to rename 3000 fiels, and packed them into one folder, the dcm2mnc can't read it at all. then we tested second way: 2. transformated DICOM to analyze. after renaming and packing 3000 dicom files, I finshed analyze transormation using [MRIcro/Import/convert foreign to analyze]. then transformated it to MINC, but the next step(Register program at BIC.MNI.McGILL) go wrong, 2 images can't be registered. But 2 above ways were nuisance because the step from GE to DICOM, so we tested 3. 3. transformate GE to MINC, after renaming and packing 3000 GE files to analyze(*.img&*.hdr) I used program ge5_to_minc at Harvard to finish it. but ge5_to_minc can't run at all. Appreciate your reply! Zhang Zhiqiang JinLing hospital, medical school of Nanjing University. P.R.China. --------------------------------- Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1?/min. From jiafucang at asisz.com Mon Jul 3 03:54:29 2006 From: jiafucang at asisz.com (Fucang Jia) Date: Mon, 3 Jul 2006 15:54:29 +0800 Subject: [MINC-users] transformation of GE to MINC References: <20060703074240.87156.qmail@web50107.mail.yahoo.com> Message-ID: <200607031554288284405@asisz.com> Hi , Zhiqiang, If you use AFNI, you can do a try with Ifile and 3dAFNItoMINC Ifile '083/I.*' '103/I.*' '123/I.*' '143/I.*' ... 3dAFNItoMINC ... HTH, Fucang On 2006-07-03 15:47:57 xingfeng lee Subject: [MINC-users] transformation of GE to MINC hello all: I encountered a problem when I transformate GE to MINC. we acquired data with GE-signa 1.5 Tesla scanner, and wanted ultimate data with MINC format. for this sake. we tested 2 ways: 1. transformate GE data to DICOM, then to MINC. At the ADW4.0 workstation of GE, exported the data of GE format to ADW to finish DICOM transformation. but there is a trouble that one folder can only hold 1000 files in GE equipment, 3000 files of one run in our experiment were divided into 3 folders, they can't be accepted by the programme of dcm2mnc(BIC MNI McGill). we tried to rename 3000 fiels, and packed them into one folder, the dcm2mnc can't read it at all. then we tested second way: 2. transformated DICOM to analyze. after renaming and packing 3000 dicom files, I finshed analyze transormation using [MRIcro/Import/convert foreign to analyze]. then transformated it to MINC, but the next step(Register program at BIC.MNI.McGILL) go wrong, 2 images can't be registered. But 2 above ways were nuisance because the step from GE to DICOM, so we tested 3. 3. transformate GE to MINC, after renaming and packing 3000 GE files to analyze(*.img&*.hdr) I used program ge5_to_minc at Harvard to finish it. but ge5_to_minc can't run at all. Appreciate your reply! Zhang Zhiqiang JinLing hospital, medical school of Nanjing University. P.R.China. --------------------------------- Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1?min. _______________________________________________ MINC-users at bic.mni.mcgill.ca http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From a.janke at gmail.com Tue Jul 4 11:17:04 2006 From: a.janke at gmail.com (Andrew Janke) Date: Tue, 4 Jul 2006 11:17:04 -0400 Subject: [MINC-users] fmr_preprocess In-Reply-To: <889df5f00606261815x1f9b0709xc41ed6098181c88e@mail.gmail.com> References: <20060626010953.25500.qmail@web50102.mail.yahoo.com> <889df5f00606260009v6bba2298ja07aad00686ee251@mail.gmail.com> <889df5f00606261815x1f9b0709xc41ed6098181c88e@mail.gmail.com> Message-ID: > So you mean the distribution you are using is Ubuntu? You can run all > software ,including fmr_preprocess, successfully on Ubuntu? what is it > version? the latest one Dapper(6.10)? At the BIC we primarilly use and support Debian sarge. (with a 2.6.8 kernel or 2.4.27). Others in the BIC are using Ubuntu but I don't know which version. a From bert at bic.mni.mcgill.ca Wed Jul 5 10:41:46 2006 From: bert at bic.mni.mcgill.ca (Robert VINCENT) Date: Wed, 5 Jul 2006 10:41:46 -0400 Subject: [MINC-users] transformation of GE to MINC In-Reply-To: <20060703074240.87156.qmail@web50107.mail.yahoo.com> Message-ID: Hi, Which version of dcm2mnc are you using? The most recent versions should easily be able to handle 3000+ input DICOM files. You can get the latest MINC sources and binaries on the packages.bic.mni.mcgill.ca site. You can give the input files by specifying the directories in which the dicom files are located: dcm2mnc input-dir-1 input-dir-2 input-dir-3 output-dir -bert On Mon, 3 Jul 2006, xingfeng lee wrote: > hello all: > I encountered a problem when I transformate GE to MINC. we acquired data with GE-signa 1.5 Tesla scanner, and wanted ultimate data with MINC format. for this sake. we tested 2 ways: > 1. transformate GE data to DICOM, then to MINC. At the ADW4.0 workstation of GE, exported the data of GE format to ADW to finish DICOM transformation. but there is a trouble that one folder can only hold 1000 files in GE equipment, 3000 files of one run in our experiment were divided into 3 folders, they can't be accepted by the programme of dcm2mnc(BIC MNI McGill). we tried to rename 3000 fiels, and packed them into one folder, the dcm2mnc can't read it at all. then we tested second way: > 2. transformated DICOM to analyze. after renaming and packing 3000 dicom files, I finshed analyze transormation using [MRIcro/Import/convert foreign to analyze]. then transformated it to MINC, but the next step(Register program at BIC.MNI.McGILL) go wrong, 2 images can't be registered. > But 2 above ways were nuisance because the step from GE to DICOM, so we tested 3. > 3. transformate GE to MINC, after renaming and packing 3000 GE files to analyze(*.img&*.hdr) I used program ge5_to_minc at Harvard to finish it. but ge5_to_minc can't run at all. > Appreciate your reply! > Zhang Zhiqiang > > JinLing hospital, medical school of Nanjing University. P.R.China. > > > > --------------------------------- > Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1?/min. > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > From sorench at gmail.com Wed Jul 5 19:22:48 2006 From: sorench at gmail.com (Soren Christensen) Date: Thu, 6 Jul 2006 09:22:48 +1000 Subject: [MINC-users] autoreg - mritoself problem Message-ID: Hi, I have a problem with mritoself in an installation that was compiled on Fedora 5: i try to co-register: mritoself fspgr.mnc DWI.mnc test.xfm That gives the output shown below. The transform file is the identity (and it should't be). Doen anyone have a clue what could have gone wrong during compilation? the only thing that seemed to cause problems during install was the MNI perlib library - the output is shown at the end of this message. Could that be the source of the problem? Thanks Soren [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] [2006-07-06 09:14:40] /usr/local/mni/bin/autocrop fspgr.mnc /usr/tmp/mritoself_12775//fspgr_crop.mnc -step -0.46875 1.5 0.46875 -extend 0,0 0,0 -50,0 -byte [autocrop] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] [2006-07-06 09:14:40] /usr/local/mni/bin/mincreshape -byte fspgr.mnc /usr/tmp/mritoself_12775//fspgr_crop.mnc -start 0,107,0 -count 120,406,512 Copying chunks:........................................................................................................................Done. [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] [2006-07-06 09:14:42] /usr/local/mni/bin/autocrop DWI.mnc /usr/tmp/mritoself_12775//DWI_crop.mnc -step -0.9375 0.93750393390655495374 5 -extend 0,0 0,0 0,0 -byte [autocrop] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] [2006-07-06 09:14:42] /usr/local/mni/bin/mincreshape -byte DWI.mnc /usr/tmp/mritoself_12775//DWI_crop.mnc -start 0,0,0 -count 20,257,256 Copying chunks:....................Done. warning: mutual information used as objective function -- target threshold forced to volume min threshold on source volume = 40 threshold on target volume = 0 [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] [2006-07-06 09:14:47] /usr/local/mni/bin/minctracc /usr/tmp/mritoself_12775//fspgr_crop.mnc /usr/tmp/mritoself_12775//DWI_crop.mnc /usr/tmp/mritoself_12775//test_tmp1.xfm -identity -est_center -lsq6 -mi -groups 256 -threshold 40 0 -step 7.3 7.3 7.3 -simplex 3 Initial objective function val = 0.00000000 Final objective function value = 0.00000000 [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] [2006-07-06 09:14:50] /usr/local/mni/bin/minctracc /usr/tmp/mritoself_12775//fspgr_crop.mnc /usr/tmp/mritoself_12775//DWI_crop.mnc test.xfm -est_center -transformation /usr/tmp/mritoself_12775//test_tmp1.xfm -lsq6 -mi -groups 256 -threshold 40 0 -step 4.3 4.3 4.3 -simplex 1.5 Initial objective function val = 0.00000000 Final objective function value = 0.00000000 The xfm file looks like: MNI Transform File %Thu Jul 6 09:14:50 2006>>> /usr/local/mni/bin/minctracc /usr/tmp/mritoself_12775//fspgr_crop.mnc /usr/tmp/mritoself_12775//DWI_crop.mnc test.xfm -est_center -transformation /usr/tmp/mritoself_12775//test_tmp1.xfm -lsq6 -mi -groups 256 -threshold 40 0 -step 4.3 4.3 4.3 -simplex 1.5 %(Package mni_autoreg 0.99.2, compiled by root at PC2-BIL (i686-pc-linux-gnu) on Tue Jul 4 13:16:13 EST 2006) Transform_Type = Linear; Linear_Transform = 1 0 0 0 0 1 0 0 0 0 1 0; MNI perllib compilation: mkdir /usr/local/mni/data perl Makefile.PL make test rm -f t/datadir.t sed 's|@DEFAULT_DATA_DIR@|/usr/local/mni/data|' t/datadir.t.in > t/datadir.t chmod 444 t/datadir.t rm -f t/toy_ls sed 's|@PERL@|\#!/usr/bin/perl|' t/toy_ls.in > t/toy_ls chmod 555 t/toy_ls PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/batch-0........ok t/batch-1......../usr/bin/getopt: invalid option -- J /usr/bin/getopt: invalid option -- Q /usr/bin/getopt: invalid option -- S /usr/bin/getopt: invalid option -- s /usr/bin/getopt: invalid option -- o /usr/bin/getopt: invalid option -- e batch: error parsing command-line arguments Error closing pipe to batch: at t/batch-1.t line 36 `batch' exited with non-zero status code: 256 at t/batch-1.t line 36 t/batch-1........NOK 1tail: cannot open `.out29303' for reading: No such file or directory t/batch-1........NOK 3/usr/bin/getopt: invalid option -- J /usr/bin/getopt: invalid option -- Q /usr/bin/getopt: invalid option -- S /usr/bin/getopt: invalid option -- s /usr/bin/getopt: invalid option -- o /usr/bin/getopt: invalid option -- e batch: error parsing command-line arguments Error closing pipe to batch: at t/batch-1.t line 36 `batch' exited with non-zero status code: 256 at t/batch-1.t line 36 t/batch-1........NOK 4tail: cannot open `.out29303' for reading: No such file or directory t/batch-1........NOK 6/usr/bin/getopt: invalid option -- J /usr/bin/getopt: invalid option -- Q /usr/bin/getopt: invalid option -- S /usr/bin/getopt: invalid option -- s /usr/bin/getopt: invalid option -- o /usr/bin/getopt: invalid option -- e batch: error parsing command-line arguments t/batch-1........NOK 8tail: cannot open `.out29303' for reading: No such file or directory t/batch-1........NOK 10/usr/bin/getopt: invalid option -- J /usr/bin/getopt: invalid option -- Q /usr/bin/getopt: invalid option -- S /usr/bin/getopt: invalid option -- s /usr/bin/getopt: invalid option -- o /usr/bin/getopt: invalid option -- e batch: error parsing command-line arguments Error closing pipe to batch: at t/batch-1.t line 87 `batch' exited with non-zero status code: 256 at t/batch-1.t line 87 t/batch-1........NOK 11tail: cannot open `.out29303' for reading: No such file or directory t/batch-1........NOK 13/usr/bin/getopt: invalid option -- J /usr/bin/getopt: invalid option -- Q /usr/bin/getopt: invalid option -- l /usr/bin/getopt: invalid option -- S /usr/bin/getopt: invalid option -- s /usr/bin/getopt: invalid option -- o /usr/bin/getopt: invalid option -- e batch: error parsing command-line arguments Error closing pipe to batch: at t/batch-1.t line 36 `batch' exited with non-zero status code: 256 at t/batch-1.t line 36 From a.janke at gmail.com Wed Jul 5 21:01:11 2006 From: a.janke at gmail.com (Andrew Janke) Date: Wed, 5 Jul 2006 21:01:11 -0400 Subject: [MINC-users] autoreg - mritoself problem In-Reply-To: References: Message-ID: Hi Soren, I suspect this is a problem with thresholding. Check that: * Your images have "sensible" ranges. * that your data is not compressed into the lowwer few bits of your data. a On 7/5/06, Soren Christensen wrote: > Hi, > I have a problem with mritoself in an installation that was compiled > on Fedora 5: > i try to co-register: > mritoself fspgr.mnc DWI.mnc test.xfm > > That gives the output shown below. The transform file is the identity > (and it should't be). > Doen anyone have a clue what could have gone wrong during compilation? > the only thing that seemed to cause problems during install was the > MNI perlib library - the output is shown at the end of this message. > Could that be the source of the problem? > > Thanks > Soren > > > > [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] > [2006-07-06 09:14:40] /usr/local/mni/bin/autocrop fspgr.mnc > /usr/tmp/mritoself_12775//fspgr_crop.mnc -step -0.46875 1.5 0.46875 > -extend 0,0 0,0 -50,0 -byte > [autocrop] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] > [2006-07-06 09:14:40] /usr/local/mni/bin/mincreshape -byte fspgr.mnc > /usr/tmp/mritoself_12775//fspgr_crop.mnc -start 0,107,0 -count > 120,406,512 > Copying chunks:........................................................................................................................Done. > [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] > [2006-07-06 09:14:42] /usr/local/mni/bin/autocrop DWI.mnc > /usr/tmp/mritoself_12775//DWI_crop.mnc -step -0.9375 > 0.93750393390655495374 5 -extend 0,0 0,0 0,0 -byte > [autocrop] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] > [2006-07-06 09:14:42] /usr/local/mni/bin/mincreshape -byte DWI.mnc > /usr/tmp/mritoself_12775//DWI_crop.mnc -start 0,0,0 -count 20,257,256 > Copying chunks:....................Done. > warning: mutual information used as objective function -- target > threshold forced to volume min > threshold on source volume = 40 > threshold on target volume = 0 > [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] > [2006-07-06 09:14:47] /usr/local/mni/bin/minctracc > /usr/tmp/mritoself_12775//fspgr_crop.mnc > /usr/tmp/mritoself_12775//DWI_crop.mnc > /usr/tmp/mritoself_12775//test_tmp1.xfm -identity -est_center -lsq6 > -mi -groups 256 -threshold 40 0 -step 7.3 7.3 7.3 -simplex 3 > Initial objective function val = 0.00000000 > Final objective function value = 0.00000000 > [mritoself] [minc at PC2-BIL:/home/minc/siewmin_temp/workbench/minc] > [2006-07-06 09:14:50] /usr/local/mni/bin/minctracc > /usr/tmp/mritoself_12775//fspgr_crop.mnc > /usr/tmp/mritoself_12775//DWI_crop.mnc test.xfm -est_center > -transformation /usr/tmp/mritoself_12775//test_tmp1.xfm -lsq6 -mi > -groups 256 -threshold 40 0 -step 4.3 4.3 4.3 -simplex 1.5 > Initial objective function val = 0.00000000 > Final objective function value = 0.00000000 > > The xfm file looks like: > MNI Transform File > %Thu Jul 6 09:14:50 2006>>> /usr/local/mni/bin/minctracc > /usr/tmp/mritoself_12775//fspgr_crop.mnc > /usr/tmp/mritoself_12775//DWI_crop.mnc test.xfm -est_center > -transformation /usr/tmp/mritoself_12775//test_tmp1.xfm -lsq6 -mi > -groups 256 -threshold 40 0 -step 4.3 4.3 4.3 -simplex 1.5 > %(Package mni_autoreg 0.99.2, compiled by root at PC2-BIL > (i686-pc-linux-gnu) on Tue Jul 4 13:16:13 EST 2006) > > Transform_Type = Linear; > Linear_Transform = > 1 0 0 0 > 0 1 0 0 > 0 0 1 0; > > > > > > > > MNI perllib compilation: > mkdir /usr/local/mni/data > perl Makefile.PL > make test > rm -f t/datadir.t > sed 's|@DEFAULT_DATA_DIR@|/usr/local/mni/data|' t/datadir.t.in > t/datadir.t > chmod 444 t/datadir.t > rm -f t/toy_ls > sed 's|@PERL@|\#!/usr/bin/perl|' t/toy_ls.in > t/toy_ls > chmod 555 t/toy_ls > PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" > "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t > t/batch-0........ok > t/batch-1......../usr/bin/getopt: invalid option -- J > /usr/bin/getopt: invalid option -- Q > /usr/bin/getopt: invalid option -- S > /usr/bin/getopt: invalid option -- s > /usr/bin/getopt: invalid option -- o > /usr/bin/getopt: invalid option -- e > batch: error parsing command-line arguments > Error closing pipe to batch: > at t/batch-1.t line 36 > `batch' exited with non-zero status code: 256 > at t/batch-1.t line 36 > > > t/batch-1........NOK 1tail: cannot open `.out29303' for reading: No > such file or directory > t/batch-1........NOK 3/usr/bin/getopt: invalid option -- J > /usr/bin/getopt: invalid option -- Q > /usr/bin/getopt: invalid option -- S > /usr/bin/getopt: invalid option -- s > /usr/bin/getopt: invalid option -- o > /usr/bin/getopt: invalid option -- e > batch: error parsing command-line arguments > Error closing pipe to batch: > at t/batch-1.t line 36 > `batch' exited with non-zero status code: 256 > at t/batch-1.t line 36 > t/batch-1........NOK 4tail: cannot open `.out29303' for reading: No > such file or directory > t/batch-1........NOK 6/usr/bin/getopt: invalid option -- J > /usr/bin/getopt: invalid option -- Q > /usr/bin/getopt: invalid option -- S > /usr/bin/getopt: invalid option -- s > /usr/bin/getopt: invalid option -- o > /usr/bin/getopt: invalid option -- e > batch: error parsing command-line arguments > t/batch-1........NOK 8tail: cannot open `.out29303' for reading: No > such file or directory > t/batch-1........NOK 10/usr/bin/getopt: invalid option -- J > /usr/bin/getopt: invalid option -- Q > /usr/bin/getopt: invalid option -- S > /usr/bin/getopt: invalid option -- s > /usr/bin/getopt: invalid option -- o > /usr/bin/getopt: invalid option -- e > batch: error parsing command-line arguments > Error closing pipe to batch: > at t/batch-1.t line 87 > `batch' exited with non-zero status code: 256 > at t/batch-1.t line 87 > t/batch-1........NOK 11tail: cannot open `.out29303' for reading: No > such file or directory > t/batch-1........NOK 13/usr/bin/getopt: invalid option -- J > /usr/bin/getopt: invalid option -- Q > /usr/bin/getopt: invalid option -- l > /usr/bin/getopt: invalid option -- S > /usr/bin/getopt: invalid option -- s > /usr/bin/getopt: invalid option -- o > /usr/bin/getopt: invalid option -- e > batch: error parsing command-line arguments > Error closing pipe to batch: > at t/batch-1.t line 36 > `batch' exited with non-zero status code: 256 > at t/batch-1.t line 36 > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From hehuiguang at gmail.com Thu Jul 6 02:44:23 2006 From: hehuiguang at gmail.com (Huiguang He) Date: Thu, 6 Jul 2006 14:44:23 +0800 Subject: [MINC-users] about minc data format conversion Message-ID: <2c8c005e0607052344y2b90e31fr7e2993a5ec74311f@mail.gmail.com> Hi, Dear Minc expert: I have the same question with minc data conversion. Our aim is try to convert GE format data to mnc file. Since ge5_to_minc doesn't work. So we use GE->DCM->Analyze->MNC, it seems stupid, but it does work. My question is : Is mnc data format big ending or little ending? Because when I try to use different tools to transfer GE->DCM, some is big ending, some is little ending. But little ending DCM file can be converted well, while big ending dcm doesn't. So, I want to confirm whether MNC data is big ending or little ending. Does anybody have some other method for transfer GE->MNC? Thanks very much! Huiguang Huiguang He, Ph.D. Chinese Academy of Sciences On 7/4/06, minc-users-request at bic.mni.mcgill.ca < minc-users-request at bic.mni.mcgill.ca> wrote: > > Send MINC-users mailing list submissions to > minc-users at bic.mni.mcgill.ca > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > or, via email, send a message with subject or body 'help' to > minc-users-request at bic.mni.mcgill.ca > > You can reach the person managing the list at > minc-users-owner at bic.mni.mcgill.ca > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of MINC-users digest..." > > > Today's Topics: > > 1. transformation of GE to MINC (xingfeng lee) > 2. Re: transformation of GE to MINC (Fucang Jia) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 3 Jul 2006 00:42:40 -0700 (PDT) > From: xingfeng lee > Subject: [MINC-users] transformation of GE to MINC > To: minc-users at bic.mni.mcgill.ca > Message-ID: <20060703074240.87156.qmail at web50107.mail.yahoo.com> > Content-Type: text/plain; charset=iso-8859-1 > > hello all: > I encountered a problem when I transformate GE to MINC. we acquired data > with GE-signa 1.5 Tesla scanner, and wanted ultimate data with MINC > format. for this sake. we tested 2 ways: > 1. transformate GE data to DICOM, then to MINC. At the ADW4.0 workstation > of GE, exported the data of GE format to ADW to finish DICOM transformation. > but there is a trouble that one folder can only hold 1000 files in GE > equipment, 3000 files of one run in our experiment were divided into 3 > folders, they can't be accepted by the programme of dcm2mnc(BIC MNI McGill). > we tried to rename 3000 fiels, and packed them into one folder, the dcm2mnc > can't read it at all. then we tested second way: > 2. transformated DICOM to analyze. after renaming and packing 3000 dicom > files, I finshed analyze transormation using [MRIcro/Import/convert foreign > to analyze]. then transformated it to MINC, but the next step(Register > program at BIC.MNI.McGILL) go wrong, 2 images can't be registered. > But 2 above ways were nuisance because the step from GE to DICOM, so we > tested 3. > 3. transformate GE to MINC, after renaming and packing 3000 GE files to > analyze(*.img&*.hdr) I used program ge5_to_minc at Harvard to finish it. but > ge5_to_minc can't run at all. > Appreciate your reply! > Zhang Zhiqiang > > JinLing hospital, medical school of Nanjing University. P.R.China. > > > > --------------------------------- > Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great > rates starting at 1?/min. > > ------------------------------ > > Message: 2 > Date: Mon, 3 Jul 2006 15:54:29 +0800 > From: "Fucang Jia" > Subject: Re: [MINC-users] transformation of GE to MINC > To: "MINC users mailing list" > Message-ID: <200607031554288284405 at asisz.com> > Content-Type: text/plain; charset="gb2312" > > Hi , Zhiqiang, > > If you use AFNI, you can do a try with Ifile and 3dAFNItoMINC > > Ifile '083/I.*' '103/I.*' '123/I.*' '143/I.*' ... > > 3dAFNItoMINC ... > > HTH, > > Fucang > > > > > On 2006-07-03 15:47:57 xingfeng lee > Subject: [MINC-users] transformation of GE to MINC > > hello all: > > I encountered a problem when I transformate GE to MINC. we acquired data with GE-signa > 1.5 > Tesla scanner, and wanted ultimate data with MINC format. for this sake. we tested 2 ways: > 1. transformate GE data to DICOM, then to MINC. At the ADW4.0 > workstation of GE, exported the data of GE format to ADW to finish DICOM transformation. but there is a trouble that one folder can only hold 1000 files in GE equipment, 3000 files of one run in our experiment were divided into 3 folders, they can't be accepted by the programme of dcm2mnc(BIC MNI McGill). we tried to rename 3000 fiels, and packed them into one folder, the dcm2mnc can't read it at all. then we tested second way: > > 2. transformated DICOM to analyze. after renaming and packing 3000 dicom files, I finshed analyze transormation using [MRIcro/Import/convert foreign to analyze]. then transformated it to MINC, but the next step(Register > program at BIC.MNI.McGILL) go wrong, 2 images can't be registered. > > But 2 above ways were nuisance because the step from GE to DICOM, so we tested 3. > > 3. transformate GE to MINC, after renaming and packing 3000 GE files to analyze(*.img&*.hdr) I used program > ge5_to_minc at Harvard > to finish it. but ge5_to_minc can't run at all. > Appreciate your reply! > Zhang Zhiqiang > > JinLing hospital, medical school of Nanjing University. P.R.China. > > > > --------------------------------- > > Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1?min. > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > ------------------------------ > > _______________________________________________ > MINC-users mailing list > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > End of MINC-users Digest, Vol 12, Issue 1 > ***************************************** > From a.janke at gmail.com Thu Jul 6 08:40:16 2006 From: a.janke at gmail.com (Andrew Janke) Date: Thu, 6 Jul 2006 08:40:16 -0400 Subject: [MINC-users] about minc data format conversion In-Reply-To: <2c8c005e0607052344y2b90e31fr7e2993a5ec74311f@mail.gmail.com> References: <2c8c005e0607052344y2b90e31fr7e2993a5ec74311f@mail.gmail.com> Message-ID: Hi Huigang, MINC itself is certainly byte swapping independent. (endianness doesn't matter). As for DICOM it is supposed to also be immune to such things, but then I can't vouch for that as I don't know all the internals of DICOM and what certain manufacturers do with certain header fields that may be affecting your conversions. If your scanner can export DICOM this nearly always turns out to be the best way to go. From there simply use dcm2mnc that is part of minc 1.4 and up. This dcm2mnc converter should also be able to convert your GE data once it has been converted to DICOM by your other converter presuming that the other converter (GE->DCM) follows standards. Andrew On 7/6/06, Huiguang He wrote: > Hi, Dear Minc expert: > I have the same question with minc data conversion. > Our aim is try to convert GE format data to mnc file. Since > ge5_to_minc doesn't work. So we use GE->DCM->Analyze->MNC, it seems stupid, > but it does work. > My question is : Is mnc data format big ending or little ending? > Because when I try to use different tools to transfer GE->DCM, some is > big ending, some is little ending. But little ending DCM file can be > converted well, while big ending dcm doesn't. > So, I want to confirm whether MNC data is big ending or little ending. > > Does anybody have some other method for transfer GE->MNC? > > Thanks very much! > Huiguang > > Huiguang He, Ph.D. > Chinese Academy of Sciences From forrest.bao at gmail.com Sun Jul 9 01:35:01 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Sun, 9 Jul 2006 13:35:01 +0800 Subject: [MINC-users] fmr_preprocess In-Reply-To: References: <20060626010953.25500.qmail@web50102.mail.yahoo.com> <889df5f00606260009v6bba2298ja07aad00686ee251@mail.gmail.com> <889df5f00606261815x1f9b0709xc41ed6098181c88e@mail.gmail.com> Message-ID: <889df5f00607082235g143cc7bfq449349371172a077@mail.gmail.com> hi guys, Now I have installed Perl and MNI modules. I has tested on both SuSE 10 and Ubuntu 6.06. But when I run fmr_preprocess, I found following problem. root at orioleQ:~/Desktop# fmr_preprocess fmr_preprocess: couldn't find program "mincblur" fmr_preprocess: couldn't find program "alignlinear" fmr_preprocess: couldn't find program "autocrop" fmr_preprocess: couldn't find program "minctracc" fmr_preprocess: couldn't find program "param2xfm" Usage: fmr_preprocess input.mnc [input2.mnc ...] -options It seems some options of fmr_preprocess doesn't work well due to lack of these programs. Such as -blur and -align. How can i fix it? Thank you for helping us. Cheers, Sheng On 7/4/06, Andrew Janke wrote: > > > So you mean the distribution you are using is Ubuntu? You can run all > > software ,including fmr_preprocess, successfully on Ubuntu? what is it > > version? the latest one Dapper(6.10)? > > At the BIC we primarilly use and support Debian sarge. (with a 2.6.8 > kernel or 2.4.27). Others in the BIC are using Ubuntu but I don't > know which version. > > > a > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- EE is no more about circuits than CS is about computers. Home Page: http://forrest.bao.googlepages.com Blog: http://blog.360.yahoo.com/forrestbao http://forrestbao.spaces.msn.com MSN: bsh1984 at hotmail.com (IM only) Gmail: forrest.bao at gmail.com (Email also) Yahoo! ID: forrestbao at yahoo.com.cn (Email also) AIM/ICQ: forrestbao Skype: forrestbao From thompson at loni.ucla.edu Sun Jul 9 02:21:46 2006 From: thompson at loni.ucla.edu (Paul Thompson) Date: Sat, 8 Jul 2006 23:21:46 -0700 Subject: [MINC-users] Using MINC on Intel Macs Message-ID: Hi Andrew + everyone - Has anyone had any luck building MINC tools on the new "Intel core" Macs, running OS 10.4 ("Tiger")? I've downloaded Andrew's Mac OS 10.4 MINC package (from the link below - thanks Andrew!) but I think it was built on a PowerPC Mac (running OS 10.4). For folks who are not used to Macs (!), Mac is now making very fast laptops that ship with Intel chips, and they can often run applications built on a Mac PowerPC (using an built-in emulator called Rosetta). I tried out Andrew's MINC package (URL is below) -- I noticed that Register and most of the main minc tools do run ok on my Intel Mac running 10.4.7 (e.g. things like mincresample run fine). However I can't link against the pre-compiled volume I/O libraries when I use gcc to build my own C code that reads MINC. gcc complains with this: /usr/bin/ld: warning /usr/local/mni/lib/libbicpl.a archive's cputype (18, architecture ppc) does not match cputype ( 7) for specified -arch flag: i386 (can't load from it) /usr/bin/ld: warning /usr/local/mni/lib/libvolume_io.a archive's cputype (18, architecture ppc) does not match cputy pe (7) for specified -arch flag: i386 (can't load from it) /usr/bin/ld: warning /usr/local/mni/lib/libminc.a archive's cputype (18, architecture ppc) does not match cputype (7 ) for specified -arch flag: i386 (can't load from it) Has anyone built these libraries on Intel Macs? If so would you be willing to post the .a files (or even email them to Andrew to put on the web? ;) Thanks a lot for any ideas! :) Paul (Paul Thompson, UCLA) http://www.loni.ucla.edu/~thompson/thompson.html ============================ >>> a.janke at gmail.com 04/05/06 10:11 PM >>> >>> Hi Paul (and others) I recompiled a bunch of them last night under 10.4 there were a few small tweaks that were needed for POSIX compliance but by en large things progressed smoothly. For the packages that require external libraries (bicpl and netpbm for example) I have used the fink versions. In the future I may use darwinports as it seems a bit "cleaner" but at the moment I am unconvinced either way. Could those who use tiger please test the new packages here: http://packages.bic.mni.mcgill.ca/osx-10.4/ Note that this is not a complete list (some failed for various reasons that I will try to rectify over the next few days). Andrew On 4/3/06, Paul Rasser wrote: > Just wondering how things are progressing regarding the Mac OS > packages? > > Paul. From forrest.bao at gmail.com Sun Jul 9 03:26:00 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Sun, 9 Jul 2006 15:26:00 +0800 Subject: [MINC-users] The Ubuntu apt-get source of MINC Message-ID: <889df5f00607090026g4a156402k9356faeb1e784900@mail.gmail.com> Hi guys, I am not sure how many of you have tried to install MINC from apt-get, if you are a Debian or Ubuntu user. This afternoon, I found accidently in "Synaptic" that MINC is available to be install via apt-get. Has anyone tried this way? Is this apt-get packed by the developers? Can I use that way to install MINC installed of downloading and compiling the source of minc and bicpl? I really don't wanna configure and compile the source code. The procedure is too complex and it cost me a lot of time. Thank you for helping me. Cheers, Sheng Bao -- EE is no more about circuits than CS is about computers. Home Page: http://forrest.bao.googlepages.com Blog: http://blog.360.yahoo.com/forrestbao http://forrestbao.spaces.msn.com MSN: bsh1984 at hotmail.com (IM only) Gmail: forrest.bao at gmail.com (Email also) Yahoo! ID: forrestbao at yahoo.com.cn (Email also) AIM/ICQ: forrestbao Skype: forrestbao From steven.robbins at videotron.ca Sun Jul 9 07:52:30 2006 From: steven.robbins at videotron.ca (Steve M. Robbins) Date: Sun, 09 Jul 2006 07:52:30 -0400 Subject: [MINC-users] The Ubuntu apt-get source of MINC In-Reply-To: <889df5f00607090026g4a156402k9356faeb1e784900@mail.gmail.com> References: <889df5f00607090026g4a156402k9356faeb1e784900@mail.gmail.com> Message-ID: <20060709115229.GG6599@nyongwa.montreal.qc.ca> On Sun, Jul 09, 2006 at 03:26:00PM +0800, Sheng Bao wrote: > This afternoon, I found accidently in "Synaptic" that MINC is available to > be install via apt-get. > Has anyone tried this way? Is this apt-get packed by the developers? The Debian packages are maintained by me. I am an alumnus of the MNI and, though never a core developer, did contribute to maintaining MINC. > Can I use that way to install MINC installed of downloading and compiling > the source of minc and bicpl? The intent is that you can install those packages rather than compiling MINC, yes. Bicpl is not included. One caveat is that currently MINC version 1.4 is packaged, and not the newer 2.0 series. I gather that version 2 is considered stable now, so I think I'll package it soon. Cheers, -Steve From jason at bic.mni.mcgill.ca Sun Jul 9 08:11:33 2006 From: jason at bic.mni.mcgill.ca (Jason Lerch) Date: Sun, 9 Jul 2006 08:11:33 -0400 Subject: [MINC-users] Using MINC on Intel Macs In-Reply-To: References: Message-ID: Greetings Paul + everyone, I have a set of the MINC tools built for intel-mac, but using MINC2 rather than MINC1. I'd certainly be happy to create an installer and distribute it if there's interest. Jason On Jul 9, 2006, at 2:21 AM, Paul Thompson wrote: > Hi Andrew + everyone - > > Has anyone had any luck building MINC tools on the new "Intel core" > Macs, running OS 10.4 ("Tiger")? > > I've downloaded Andrew's Mac OS 10.4 MINC package (from the link > below - thanks Andrew!) but I think it was built on a PowerPC > Mac (running OS 10.4). For folks who are not used to Macs (!), Mac is > now making very fast laptops that ship with Intel chips, > and they can often run applications built on a Mac PowerPC (using an > built-in emulator called Rosetta). I tried out > Andrew's MINC package (URL is below) -- I noticed that Register and > most of the main minc tools do run ok on my > Intel Mac running 10.4.7 (e.g. things like mincresample run fine). > > However I can't link against the pre-compiled volume I/O libraries > when I use gcc to build my own C code that reads MINC. gcc > complains with this: > > /usr/bin/ld: warning /usr/local/mni/lib/libbicpl.a archive's cputype > (18, architecture ppc) does not match cputype ( > 7) for specified -arch flag: i386 (can't load from it) > /usr/bin/ld: warning /usr/local/mni/lib/libvolume_io.a archive's > cputype (18, architecture ppc) does not match cputy > pe (7) for specified -arch flag: i386 (can't load from it) > /usr/bin/ld: warning /usr/local/mni/lib/libminc.a archive's cputype > (18, architecture ppc) does not match cputype (7 > ) for specified -arch flag: i386 (can't load from it) > > Has anyone built these libraries on Intel Macs? If so would you be > willing to post the .a files (or even email them to Andrew to put on > the web? ;) > > Thanks a lot for any ideas! :) > Paul > (Paul Thompson, UCLA) > http://www.loni.ucla.edu/~thompson/thompson.html > > > ============================ >>>> a.janke at gmail.com 04/05/06 10:11 PM >>> >>>> > Hi Paul (and others) > > I recompiled a bunch of them last night under 10.4 there were a few > small tweaks that were needed for POSIX compliance but by en large > things progressed smoothly. For the packages that require external > libraries (bicpl and netpbm for example) I have used the fink > versions. In the future I may use darwinports as it seems a bit > "cleaner" but at the moment I am unconvinced either way. > > Could those who use tiger please test the new packages here: > > http://packages.bic.mni.mcgill.ca/osx-10.4/ > > Note that this is not a complete list (some failed for various reasons > that I will try to rectify over the next few days). > > > Andrew > > > On 4/3/06, Paul Rasser wrote: > >> Just wondering how things are progressing regarding the Mac OS >> packages? >> >> Paul. > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From a.janke at gmail.com Sun Jul 9 23:39:24 2006 From: a.janke at gmail.com (Andrew Janke) Date: Sun, 9 Jul 2006 23:39:24 -0400 Subject: [MINC-users] The Ubuntu apt-get source of MINC In-Reply-To: <20060709115229.GG6599@nyongwa.montreal.qc.ca> References: <889df5f00607090026g4a156402k9356faeb1e784900@mail.gmail.com> <20060709115229.GG6599@nyongwa.montreal.qc.ca> Message-ID: Hi Sheng, Your other option is to add packages.bic.mni.mcgill.ca/deb to your sources.list The exact line to add is at the bottom of the link above. a On 7/9/06, Steve M. Robbins wrote: > On Sun, Jul 09, 2006 at 03:26:00PM +0800, Sheng Bao wrote: > > > This afternoon, I found accidently in "Synaptic" that MINC is available to > > be install via apt-get. > > Has anyone tried this way? Is this apt-get packed by the developers? > > The Debian packages are maintained by me. I am an alumnus of the MNI > and, though never a core developer, did contribute to maintaining > MINC. > > > > Can I use that way to install MINC installed of downloading and compiling > > the source of minc and bicpl? > > The intent is that you can install those packages rather than > compiling MINC, yes. Bicpl is not included. > > One caveat is that currently MINC version 1.4 is packaged, and not the > newer 2.0 series. I gather that version 2 is considered stable now, > so I think I'll package it soon. > > Cheers, > -Steve > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From forrest.bao at gmail.com Mon Jul 10 05:48:37 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Mon, 10 Jul 2006 17:48:37 +0800 Subject: [MINC-users] where does these programs contain? Message-ID: <889df5f00607100248i6ff6adb4sdd4ddf95f9fdee9c@mail.gmail.com> Hi guys, I have installed MINC and BICPL but when I tries to run fmr_preprocess, I found some programs can't be found. I have specified the location to install programs by the --prefix arguement. And I can't find them at /usr/local/mni/bin So, can anyone tell me which packages contain these programs? These programs are : mincblur, alignnear, autocrop, minctracc, param2xfm root at orioleQ:~/fmri/bicpl-1.4.3# fmr_preprocess fmr_preprocess: couldn't find program "mincblur" fmr_preprocess: couldn't find program "alignlinear" fmr_preprocess: couldn't find program "autocrop" fmr_preprocess: couldn't find program "minctracc" fmr_preprocess: couldn't find program "param2xfm" Usage: fmr_preprocess input.mnc [input2.mnc ...] -options Thank you for helping me. Cheers, Sheng -- EE is no more about circuits than CS is about computers. Home Page: http://forrest.bao.googlepages.com Blog: http://blog.360.yahoo.com/forrestbao http://forrestbao.spaces.msn.com MSN: bsh1984 at hotmail.com (IM only) Gmail: forrest.bao at gmail.com (Email also) Yahoo! ID: forrestbao at yahoo.com.cn (Email also) AIM/ICQ: forrestbao Skype: forrestbao From forrest.bao at gmail.com Mon Jul 10 08:04:45 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Mon, 10 Jul 2006 20:04:45 +0800 Subject: [MINC-users] AIR installation problem Message-ID: <889df5f00607100504g638f6ba1w261f831eaa70caa2@mail.gmail.com> Hi guys, I have solved all problems of using fmr_preprocess except "alignlinear". (So please discard two of my previous email.) I searched on the internet and found "alignlinear" is contained in a package named AIR(Automated Image Registration). So I download it from LONI. And I encountered a problem when installed it. I extract the tar.gz ball but found no configure script. root at orioleQ:~/fmri/AIR3.08# ls -a . alignmritopet.com binarymask.c FILELIST.h LICENSE.h reconcile_air.c reslice_warp.tcl sizeof.c .. alignpettomri.com binarymath.c fixheader.c magnify.c reconcile_rigidbody.c reunite.c softmean.c AIRmain.h alignpettopet.com cd_air.c gsmooth.c makeaheader.c reorient.c scanair.c src AIRnlmain.h align.tcl cd_warp.c HEADERmain.h MAKE.com resize.c scanheader.c VERSION.h alignlinear.c align_warp.c combine_air.c identify.c manualreslice.c reslice.c scan_warp.c zoomer.c alignmritoatlas.com align_warp.tcl combine_warp.c invert_air.c mv_air.c reslice.tcl separate.c alignmritomri.com binarize.c definecommon_air.c layout.c mv_warp.c reslice_warp.c setheadermax.c I don't know whether I downloaded correctly. If you have a workable source code tar.gz ball, could you please tell me how to obtain it? Or email to me? Cheers, Sheng -- EE is no more about circuits than CS is about computers. Home Page: http://forrest.bao.googlepages.com Blog: http://blog.360.yahoo.com/forrestbao http://forrestbao.spaces.msn.com MSN: bsh1984 at hotmail.com (IM only) Gmail: forrest.bao at gmail.com (Email also) Yahoo! ID: forrestbao at yahoo.com.cn (Email also) AIM/ICQ: forrestbao Skype: forrestbao From forrest.bao at gmail.com Mon Jul 10 08:21:35 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Mon, 10 Jul 2006 20:21:35 +0800 Subject: [MINC-users] all problem Ok Message-ID: <889df5f00607100521i74c23794x40fef211ad6e698f@mail.gmail.com> Hi guys, I have handle everything successfully, including "alignlinear". Please forgive me for I have send three emails. I found the problem was due to AIR 3 version don't have a configure script but it is contained in AIR 5 version. Funny. Cheers, Sheng -- EE is no more about circuits than CS is about computers. Home Page: http://forrest.bao.googlepages.com Blog: http://blog.360.yahoo.com/forrestbao http://forrestbao.spaces.msn.com MSN: bsh1984 at hotmail.com (IM only) Gmail: forrest.bao at gmail.com (Email also) Yahoo! ID: forrestbao at yahoo.com.cn (Email also) AIM/ICQ: forrestbao Skype: forrestbao From jharlap at bic.mni.mcgill.ca Mon Jul 10 08:28:50 2006 From: jharlap at bic.mni.mcgill.ca (Jonathan Harlap) Date: Mon, 10 Jul 2006 08:28:50 -0400 Subject: [MINC-users] where does these programs contain? In-Reply-To: <889df5f00607100248i6ff6adb4sdd4ddf95f9fdee9c@mail.gmail.com> References: <889df5f00607100248i6ff6adb4sdd4ddf95f9fdee9c@mail.gmail.com> Message-ID: <5D3BE852-E2AF-4982-9E58-E8C0205AF420@bic.mni.mcgill.ca> Hi Sheng, These programs are part of the mni_autoreg package. Cheers, J On 10-Jul-06, at 5:48 AM, Sheng Bao wrote: > Hi guys, I have installed MINC and BICPL but when I tries to run > fmr_preprocess, I found some programs can't be found. I have > specified the > location to install programs by the --prefix arguement. And I can't > find > them at /usr/local/mni/bin > > So, can anyone tell me which packages contain these programs? These > programs > are : mincblur, alignnear, autocrop, minctracc, param2xfm > > root at orioleQ:~/fmri/bicpl-1.4.3# fmr_preprocess > fmr_preprocess: couldn't find program "mincblur" > fmr_preprocess: couldn't find program "alignlinear" > fmr_preprocess: couldn't find program "autocrop" > fmr_preprocess: couldn't find program "minctracc" > fmr_preprocess: couldn't find program "param2xfm" > > Usage: fmr_preprocess input.mnc [input2.mnc ...] -options > > Thank you for helping me. > > Cheers, > Sheng > > -- > EE is no more about circuits than CS is about computers. > > Home Page: http://forrest.bao.googlepages.com > Blog: http://blog.360.yahoo.com/forrestbao > http://forrestbao.spaces.msn.com > MSN: bsh1984 at hotmail.com (IM only) > Gmail: forrest.bao at gmail.com (Email also) > Yahoo! ID: forrestbao at yahoo.com.cn (Email also) > AIM/ICQ: forrestbao > Skype: forrestbao > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From duchesne at ieee.org Mon Jul 10 15:32:20 2006 From: duchesne at ieee.org (Simon Duchesne) Date: Mon, 10 Jul 2006 21:32:20 +0200 Subject: [MINC-users] Using MINC on Intel Macs In-Reply-To: Message-ID: Paul Andrew has put a MINC package built on/for Intel MACs. See the distribs on packages.bic.mni.mcgill.ca This being said, there's a few bugs on install; quite possibly rebuilding netcdf-mni will fix them. Later, Simon On 9/07/06 14:11, "Jason Lerch" wrote: > Greetings Paul + everyone, > > I have a set of the MINC tools built for intel-mac, but using MINC2 > rather than MINC1. I'd certainly be happy to create an installer and > distribute it if there's interest. > > Jason > > On Jul 9, 2006, at 2:21 AM, Paul Thompson wrote: > >> Hi Andrew + everyone - >> >> Has anyone had any luck building MINC tools on the new "Intel core" >> Macs, running OS 10.4 ("Tiger")? >> >> I've downloaded Andrew's Mac OS 10.4 MINC package (from the link >> below - thanks Andrew!) but I think it was built on a PowerPC >> Mac (running OS 10.4). For folks who are not used to Macs (!), Mac is >> now making very fast laptops that ship with Intel chips, >> and they can often run applications built on a Mac PowerPC (using an >> built-in emulator called Rosetta). I tried out >> Andrew's MINC package (URL is below) -- I noticed that Register and >> most of the main minc tools do run ok on my >> Intel Mac running 10.4.7 (e.g. things like mincresample run fine). >> >> However I can't link against the pre-compiled volume I/O libraries >> when I use gcc to build my own C code that reads MINC. gcc >> complains with this: >> >> /usr/bin/ld: warning /usr/local/mni/lib/libbicpl.a archive's cputype >> (18, architecture ppc) does not match cputype ( >> 7) for specified -arch flag: i386 (can't load from it) >> /usr/bin/ld: warning /usr/local/mni/lib/libvolume_io.a archive's >> cputype (18, architecture ppc) does not match cputy >> pe (7) for specified -arch flag: i386 (can't load from it) >> /usr/bin/ld: warning /usr/local/mni/lib/libminc.a archive's cputype >> (18, architecture ppc) does not match cputype (7 >> ) for specified -arch flag: i386 (can't load from it) >> >> Has anyone built these libraries on Intel Macs? If so would you be >> willing to post the .a files (or even email them to Andrew to put on >> the web? ;) >> >> Thanks a lot for any ideas! :) >> Paul >> (Paul Thompson, UCLA) >> http://www.loni.ucla.edu/~thompson/thompson.html >> >> >> ============================ >>>>> a.janke at gmail.com 04/05/06 10:11 PM >>> >>>>> >> Hi Paul (and others) >> >> I recompiled a bunch of them last night under 10.4 there were a few >> small tweaks that were needed for POSIX compliance but by en large >> things progressed smoothly. For the packages that require external >> libraries (bicpl and netpbm for example) I have used the fink >> versions. In the future I may use darwinports as it seems a bit >> "cleaner" but at the moment I am unconvinced either way. >> >> Could those who use tiger please test the new packages here: >> >> http://packages.bic.mni.mcgill.ca/osx-10.4/ >> >> Note that this is not a complete list (some failed for various reasons >> that I will try to rectify over the next few days). >> >> >> Andrew >> >> >> On 4/3/06, Paul Rasser wrote: >> >>> Just wondering how things are progressing regarding the Mac OS >>> packages? >>> >>> Paul. >> _______________________________________________ >> MINC-users at bic.mni.mcgill.ca >> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users Simon Duchesne, ing., Ph.D. Post-doctoral researcher Chercheur post-doctoral ---------------------------------------------------- Unit?/Projet VisAGeS U746 INSERM/INRIA/CNRS/U. de Rennes I IRISA Campus de Beaulieu, 35042 Rennes Cedex, France Ph: +33 (2) 99 84 25 23 / Fax: +33 (2) 99 84 71 71 INSERM Universite de Rennes I, CS 34317, 35043 Rennes Cedex, France Ph: +33 (2) 23 23 45 88 / Fax: +33 (2) 99 84 71 71 ----------------------------------------------------- duchesne at ieee.org From atsuko at bic.mni.mcgill.ca Mon Jul 10 15:45:58 2006 From: atsuko at bic.mni.mcgill.ca (Atsuko Nagano) Date: Tue, 11 Jul 2006 04:45:58 +0900 Subject: [MINC-users] coordinate question In-Reply-To: Message-ID: <000801c6a459$7afa33a0$a7b2ce84@quark> Hello all, If a voxel coordinate is [x,y,z] in a native space, and the transfer matrix is example.xfm, how to get the coordinate after the transfer using a command in UNIX? Best, Atsuko Nagano-Saito From a.janke at gmail.com Mon Jul 10 16:16:22 2006 From: a.janke at gmail.com (Andrew Janke) Date: Mon, 10 Jul 2006 16:16:22 -0400 Subject: [MINC-users] coordinate question In-Reply-To: <000801c6a459$7afa33a0$a7b2ce84@quark> References: <000801c6a459$7afa33a0$a7b2ce84@quark> Message-ID: Hi Atsuko, I suspect you will want to do something like this: 1. First get the world co-ordinate using 'voxeltoworld' (the order of the voxel co-ordinate will depend upon the MINC file order). 2. transform your world co-ordinate using something like octave. I don't know of a MINC specific command that will transform an input point from the C/L. The other option is to convert your world co-ordinate to a tagfile and then transform the tag file. Andrew On 7/10/06, Atsuko Nagano wrote: > Hello all, > > If a voxel coordinate is [x,y,z] in a native space, and the transfer > matrix is example.xfm, how to get the coordinate after the transfer > using a command in UNIX? > > Best, > Atsuko Nagano-Saito > > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From a.janke at gmail.com Mon Jul 10 16:19:40 2006 From: a.janke at gmail.com (Andrew Janke) Date: Mon, 10 Jul 2006 16:19:40 -0400 Subject: [MINC-users] Using MINC on Intel Macs In-Reply-To: References: Message-ID: On 7/10/06, Simon Duchesne wrote: > Andrew has put a MINC package built on/for Intel MACs. See the distribs on > packages.bic.mni.mcgill.ca Correct, I put what I had done here: http://packages.bic.mni.mcgill.ca/osx-10.4-intel/ > This being said, there's a few bugs on install; quite possibly rebuilding > netcdf-mni will fix them. This too is correct. However I no longer have (immediate) access to an intel based mac to rebuild the required packages! :( Until apple makes nice and gives me one I'm afraid I'm a bit stymied. a From a.janke at gmail.com Mon Jul 10 16:33:27 2006 From: a.janke at gmail.com (Andrew Janke) Date: Mon, 10 Jul 2006 16:33:27 -0400 Subject: [MINC-users] Bug fix release of ebtks (and thus N3) Message-ID: Hi all, If you have even seen a message along these lines when running N3: inputCompactField(): Incorrect number of coefficients Failure reading field file: blah-tiddly-blah.imp then this message is for you. This was caused by a "misunderstanding" between ebtks and n3 as to how to define the number of nodes in a particular datatype. Of course it was only manifest in a few special cases. So if you have been affected by this bug, your results will not change, N3 will now just start to work on these files! There is now a bug-fix release of ebtks here: http://packages.bic.mni.mcgill.ca/tgz/ebtks-1.6.1.tar.gz And some binary packages here for n3 for debian linux. http://packages.bic.mni.mcgill.ca/deb/ebtks-1.6.1-linux-2.6-intel.deb http://packages.bic.mni.mcgill.ca/deb/n3-1.10.1-linux-2.6-intel.deb The other platform updates will follow slowly. :) -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From atsuko at bic.mni.mcgill.ca Mon Jul 10 16:47:08 2006 From: atsuko at bic.mni.mcgill.ca (Atsuko Nagano) Date: Tue, 11 Jul 2006 05:47:08 +0900 Subject: [MINC-users] coordinate question In-Reply-To: Message-ID: <000901c6a462$08efa070$a7b2ce84@quark> Thank you Andrew, Transform_tags seems to work well. I will use the command. Atsuko -----Original Message----- From: minc-users-bounces at bic.mni.mcgill.ca [mailto:minc-users-bounces at bic.mni.mcgill.ca] On Behalf Of Andrew Janke Sent: Tuesday, July 11, 2006 5:16 AM To: MINC users mailing list Subject: Re: [MINC-users] coordinate question Hi Atsuko, I suspect you will want to do something like this: 1. First get the world co-ordinate using 'voxeltoworld' (the order of the voxel co-ordinate will depend upon the MINC file order). 2. transform your world co-ordinate using something like octave. I don't know of a MINC specific command that will transform an input point from the C/L. The other option is to convert your world co-ordinate to a tagfile and then transform the tag file. Andrew On 7/10/06, Atsuko Nagano wrote: > Hello all, > > If a voxel coordinate is [x,y,z] in a native space, and the transfer > matrix is example.xfm, how to get the coordinate after the transfer > using a command in UNIX? > > Best, > Atsuko Nagano-Saito > > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 _______________________________________________ MINC-users at bic.mni.mcgill.ca http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From steven.robbins at videotron.ca Mon Jul 10 17:50:41 2006 From: steven.robbins at videotron.ca (Steve M. Robbins) Date: Mon, 10 Jul 2006 17:50:41 -0400 Subject: [MINC-users] coordinate question In-Reply-To: References: <000801c6a459$7afa33a0$a7b2ce84@quark> Message-ID: <20060710215041.GJ6599@nyongwa.montreal.qc.ca> On Mon, Jul 10, 2006 at 04:16:22PM -0400, Andrew Janke wrote: > 2. transform your world co-ordinate using something like octave. > > > I don't know of a MINC specific command that will transform an input > point from the C/L. The other option is to convert your world > co-ordinate to a tagfile and then transform the tag file. Here's a simple program I wrote years ago that transforms points. Do with it what you like. -Steve /* Apply an MNI general transform to one or more points. */ // TODO: add facility to invert transform #include #include extern "C" { # include } #include using namespace std; /* Read one point per line from in, transform, and * write on out. */ void transform_points( std::istream& in, General_transform& transform, std::ostream& out ) { while( !in.eof() ) { Real x,y,z; in >> x; in >> y; in >> z; if ( in.eof() ) break; Real tx,ty,tz; general_transform_point( &transform, x, y, z, &tx, &ty, &tz ); out << tx << " " << ty << " " << tz << endl; } } const char* transform_type_name[] = { "linear", "thin plate spline", "user-defined", "concatenated", "grid" }; /* Diagnostics */ void display_transform_info( std::ostream& os, General_transform* t, std::string prefix = "" ) { os << prefix << "Type of transform: "; if ( t->type > GRID_TRANSFORM ) { os << "type " << t->type << " [UNKNOWN]"; } else { os << transform_type_name[t->type]; } os << endl << prefix << "Invert flag: " << t->inverse_flag << endl; if ( t->type == CONCATENATED_TRANSFORM ) { int n = get_n_concated_transforms(t); for( int i = 0; i < n; ++i ) { display_transform_info( os, get_nth_general_transform(t,i), prefix + " " ); } } } int main( int ac, char* av[] ) { static int verbose = 0; ArgvInfo argTable[] = { { "-verbose", ARGV_CONSTANT, (char*)1, (char*)&verbose, "emit diagnostic info" }, { NULL, ARGV_END, NULL, NULL, NULL } }; if ( ParseArgv( &ac, av, argTable, 0 ) || ac != 2 ) { cerr << "usage: " << av[0] << " transformation" << endl; return 1; } General_transform transform; if ( input_transform_file( av[1], &transform ) != OK ) { cerr << av[0] << ": cannot read transform file: " << av[1] << endl; return 2; } if (verbose) display_transform_info( std::cerr, &transform ); transform_points( std::cin, transform, std::cout ); return 0; } From thompson at loni.ucla.edu Mon Jul 10 19:47:37 2006 From: thompson at loni.ucla.edu (Paul Thompson) Date: Mon, 10 Jul 2006 18:47:37 -0500 Subject: [MINC-users] Using MINC on Intel Macs In-Reply-To: References: Message-ID: <997DCA7B-DF04-46A7-9D38-1C74D3986D21@loni.ucla.edu> Andrew, Simon, Jason -- thanks to all of you, these are much appreciated! I am having a go with them now, and they do seem to be working, except for libvolume_io.a, which I'm having a look at. See you (we'll have to pitch in money to buy Andrew an Intel Mac ;) Paul http://www.loni.ucla.edu/~thompson/thompson.html On Jul 10, 2006, at 3:19 PM, Andrew Janke wrote: > On 7/10/06, Simon Duchesne wrote: > >> Andrew has put a MINC package built on/for Intel MACs. See the >> distribs on >> packages.bic.mni.mcgill.ca > > Correct, I put what I had done here: > > http://packages.bic.mni.mcgill.ca/osx-10.4-intel/ > >> This being said, there's a few bugs on install; quite possibly >> rebuilding >> netcdf-mni will fix them. > > This too is correct. However I no longer have (immediate) access to > an intel based mac to rebuild the required packages! :( Until apple > makes nice and gives me one I'm afraid I'm a bit stymied. > > > a > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From ed.gronenschild at mi.unimaas.nl Tue Jul 11 05:51:18 2006 From: ed.gronenschild at mi.unimaas.nl (Ed Gronenschild) Date: Tue, 11 Jul 2006 11:51:18 +0200 Subject: [MINC-users] How to create colored label surface overlay Message-ID: <369F2F66-E425-4A9E-ABE9-E5B21A634416@mi.unimaas.nl> Hi, I have an .obj file holding a 3D surface of a skullstripped brain. With Display I've labeled a certain brain structure which I would like to view as a surface overlayed on the existing obj shown by Display and in the same color as the label itself. How can I do that? Ed From jason at bic.mni.mcgill.ca Tue Jul 11 07:56:59 2006 From: jason at bic.mni.mcgill.ca (Jason Lerch) Date: Tue, 11 Jul 2006 07:56:59 -0400 Subject: [MINC-users] How to create colored label surface overlay In-Reply-To: <369F2F66-E425-4A9E-ABE9-E5B21A634416@mi.unimaas.nl> References: <369F2F66-E425-4A9E-ABE9-E5B21A634416@mi.unimaas.nl> Message-ID: Use the command line program set_object_colour and then use Display two load both obj files. Alternately, ray_trace can give you prettier output. If the labelled structure is inside the surface, the command set_object_opacity might also come in handy. The two set_* commands are in conglomerate, I believe. Jason On 11-Jul-06, at 5:51 AM, Ed Gronenschild wrote: > Hi, > > I have an .obj file holding a 3D surface of a skullstripped brain. > With Display I've labeled a certain brain structure which I would > like to view as a surface overlayed on the existing obj shown by > Display and in the same color as the label itself. > How can I do that? > > Ed > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From jason at bic.mni.mcgill.ca Tue Jul 11 09:44:09 2006 From: jason at bic.mni.mcgill.ca (Jason Lerch) Date: Tue, 11 Jul 2006 09:44:09 -0400 Subject: [MINC-users] Using MINC on Intel Macs In-Reply-To: <997DCA7B-DF04-46A7-9D38-1C74D3986D21@loni.ucla.edu> References: <997DCA7B-DF04-46A7-9D38-1C74D3986D21@loni.ucla.edu> Message-ID: <44B3AB29.6090708@bic.mni.mcgill.ca> A full installer (370Mb) can be found at: http://www.bic.mni.mcgill.ca/users/jason/mactel/ This will install into /usr/local/minc2 (well, it's supposed to, so let me know if it decides to put the software somewhere else). Everything is compiled against a modified version of minc 2.0.09. Let me know of any bugs anyone finds. And if you do, do so quickly - I too will lose easy access to an intel mac by the end of the week. Known bugs: * bicpl not compiled against an image library (i.e. ray_trace would not work compiled against this version of bicpl) * perl libraries have to be installed separately. Jason Paul Thompson wrote: >Andrew, Simon, Jason -- thanks to all of you, these are much >appreciated! I am having a go with them now, and they do seem to be >working, except for libvolume_io.a, which I'm having a look at. >See you (we'll have to pitch in money to buy Andrew an Intel Mac ;) >Paul >http://www.loni.ucla.edu/~thompson/thompson.html > >On Jul 10, 2006, at 3:19 PM, Andrew Janke wrote: > > > >>On 7/10/06, Simon Duchesne wrote: >> >> >> >>>Andrew has put a MINC package built on/for Intel MACs. See the >>>distribs on >>>packages.bic.mni.mcgill.ca >>> >>> >>Correct, I put what I had done here: >> >> http://packages.bic.mni.mcgill.ca/osx-10.4-intel/ >> >> >> >>>This being said, there's a few bugs on install; quite possibly >>>rebuilding >>>netcdf-mni will fix them. >>> >>> >>This too is correct. However I no longer have (immediate) access to >>an intel based mac to rebuild the required packages! :( Until apple >>makes nice and gives me one I'm afraid I'm a bit stymied. >> >> >>a >>_______________________________________________ >>MINC-users at bic.mni.mcgill.ca >>http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users >> >> > >_______________________________________________ >MINC-users at bic.mni.mcgill.ca >http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > From tpb1 at ualberta.ca Tue Jul 11 14:53:20 2006 From: tpb1 at ualberta.ca (tpb1@ualberta.ca) Date: Tue, 11 Jul 2006 12:53:20 -0600 Subject: [MINC-users] How to create colored label surface overlay Message-ID: <20060711125320.ry73z06i04s4o4gs@webmail.ualberta.ca> Hi Ed, As an alternative to Jason's command line suggestions, I've been able to overlay an obj with its particular label colour in Display by the menu item Colour Coding --> Colour Code object, when the cursor is over top of the label of interest. You can change the label opacity in this same menu. Thomas Date: Tue, 11 Jul 2006 11:51:18 +0200 From: Ed Gronenschild Subject: [MINC-users] How to create colored label surface overlay To: minc-users at bic.mni.mcgill.ca Message-ID: <369F2F66-E425-4A9E-ABE9-E5B21A634416 at mi.unimaas.nl> Content-Type: text/plain; charset=US-ASCII; format=flowed Hi, I have an .obj file holding a 3D surface of a skullstripped brain. With Display I've labeled a certain brain structure which I would like to view as a surface overlayed on the existing obj shown by Display and in the same color as the label itself. How can I do that? Ed From a.janke at gmail.com Wed Jul 12 08:26:00 2006 From: a.janke at gmail.com (Andrew Janke) Date: Wed, 12 Jul 2006 08:26:00 -0400 Subject: [MINC-users] fmr_preprocess In-Reply-To: <889df5f00607082235g143cc7bfq449349371172a077@mail.gmail.com> References: <20060626010953.25500.qmail@web50102.mail.yahoo.com> <889df5f00606260009v6bba2298ja07aad00686ee251@mail.gmail.com> <889df5f00606261815x1f9b0709xc41ed6098181c88e@mail.gmail.com> <889df5f00607082235g143cc7bfq449349371172a077@mail.gmail.com> Message-ID: Sheng, Most of these programs are part of mni-autoreg. a On 7/9/06, Sheng Bao wrote: > hi guys, > Now I have installed Perl and MNI modules. I has tested on both SuSE 10 and > Ubuntu 6.06. > But when I run fmr_preprocess, I found following problem. > > root at orioleQ:~/Desktop# fmr_preprocess > fmr_preprocess: couldn't find program "mincblur" > fmr_preprocess: couldn't find program "alignlinear" > fmr_preprocess: couldn't find program "autocrop" > fmr_preprocess: couldn't find program "minctracc" > fmr_preprocess: couldn't find program "param2xfm" > > Usage: fmr_preprocess input.mnc [input2.mnc ...] -options > > It seems some options of fmr_preprocess doesn't work well due to lack of > these programs. Such as -blur and -align. > > How can i fix it? > > Thank you for helping us. > > Cheers, > Sheng > > On 7/4/06, Andrew Janke wrote: > > > > > So you mean the distribution you are using is Ubuntu? You can run all > > > software ,including fmr_preprocess, successfully on Ubuntu? what is it > > > version? the latest one Dapper(6.10)? > > > > At the BIC we primarilly use and support Debian sarge. (with a 2.6.8 > > kernel or 2.4.27). Others in the BIC are using Ubuntu but I don't > > know which version. > > > > > > a > > _______________________________________________ > > MINC-users at bic.mni.mcgill.ca > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > > > > -- > EE is no more about circuits than CS is about computers. > > Home Page: http://forrest.bao.googlepages.com > Blog: http://blog.360.yahoo.com/forrestbao > http://forrestbao.spaces.msn.com > MSN: bsh1984 at hotmail.com (IM only) > Gmail: forrest.bao at gmail.com (Email also) > Yahoo! ID: forrestbao at yahoo.com.cn (Email also) > AIM/ICQ: forrestbao > Skype: forrestbao > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From tpb1 at ualberta.ca Thu Jul 13 14:41:02 2006 From: tpb1 at ualberta.ca (tpb1@ualberta.ca) Date: Thu, 13 Jul 2006 12:41:02 -0600 Subject: [MINC-users] Using MINC on Intel Macs In-Reply-To: References: Message-ID: <20060713124102.bqh3ro5ugccocos4@webmail.ualberta.ca> Hi Jason, I've been able to use Display from your mactel download on my Intel iMac (running Tiger) without a hitch. I just had to make sure I added /usr/local/minc2 to my path. Thanks for making it available! TB > ------------------------------ > > Message: 3 > Date: Tue, 11 Jul 2006 09:44:09 -0400 > From: Jason Lerch > Subject: Re: [MINC-users] Using MINC on Intel Macs > To: MINC users mailing list > Message-ID: <44B3AB29.6090708 at bic.mni.mcgill.ca> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > A full installer (370Mb) can be found at: > > http://www.bic.mni.mcgill.ca/users/jason/mactel/ > > This will install into /usr/local/minc2 (well, it's supposed to, so let > me know if it decides to put the software somewhere else). Everything is > compiled against a modified version of minc 2.0.09. > > Let me know of any bugs anyone finds. And if you do, do so quickly - I > too will lose easy access to an intel mac by the end of the week. > > Known bugs: > * bicpl not compiled against an image library (i.e. ray_trace would not > work compiled against this version of bicpl) > * perl libraries have to be installed separately. > > Jason From zhangzq2001 at 126.com Fri Jul 14 08:13:53 2006 From: zhangzq2001 at 126.com (zhangzq2001) Date: Fri, 14 Jul 2006 20:13:53 +0800 (CST) Subject: [MINC-users] transformation of GE to MINC Message-ID: <44B78A81.000106.26615@m64.126.com> Hello all: I encountered a problem when I transformate GE to MINC. we acquired data with GE-signa 1.5 Tesla scanner, and wanted ultimate data with MINC format. for this sake. we tested 3 ways: 1. transformate GE data to DICOM, then to MINC. At the ADW4.0 workstation of GE, exported the data of GE format to ADW to finish DICOM transformation. but there is a trouble that one folder can only hold 1000 files in GE equipment, 3000 files of one run in our experiment were divided into 3 folder, they can't be accepted by the programme of dcm2mnc(BIC MNI McGill). we tried to rename 3000 fiels, and packed them into one folder, the dcm2mnc can't read it at all. then we tested the second way: 2. transformated DICOM to analyze. after renaming and packing 3000 dicom files, I finshed analyze transormation using [MRIcro/Import/convert foreign to analyze]. then transformated it to MINC, but next step(Register program at BIC.MNI.McGILL) go wrong, 2 images can't be registered. But 2 above ways were nuisance because the step from GE to DICOM, so we tested 3. 3. transformate GE to MINC, after renaming and packing 3000 GE files to analyze(*.img&*.hdr) I used program ge5_to_minc to finish it. but ge5_to_minc can't run at all, it displayed: "1024 error". Then, what should I do? Appreciate any reply of yours! _________________________ Zhang Zhiqiang JinLing hospital, medical school of Nanjing University. P.R.China. -- ??? ??????? From forrest.bao at gmail.com Sat Jul 15 22:27:29 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Sun, 16 Jul 2006 10:27:29 +0800 Subject: [MINC-users] fmr_preprocess In-Reply-To: References: <20060626010953.25500.qmail@web50102.mail.yahoo.com> <889df5f00606260009v6bba2298ja07aad00686ee251@mail.gmail.com> <889df5f00606261815x1f9b0709xc41ed6098181c88e@mail.gmail.com> <889df5f00607082235g143cc7bfq449349371172a077@mail.gmail.com> Message-ID: <889df5f00607151927uadbbee4j80e0890e75a72da7@mail.gmail.com> yes, I have handle everything. And they all work well now. Thank you very much. On 7/12/06, Andrew Janke wrote: > > Sheng, > > Most of these programs are part of mni-autoreg. > > a > > On 7/9/06, Sheng Bao wrote: > > hi guys, > > Now I have installed Perl and MNI modules. I has tested on both SuSE 10 > and > > Ubuntu 6.06. > > But when I run fmr_preprocess, I found following problem. > > > > root at orioleQ:~/Desktop# fmr_preprocess > > fmr_preprocess: couldn't find program "mincblur" > > fmr_preprocess: couldn't find program "alignlinear" > > fmr_preprocess: couldn't find program "autocrop" > > fmr_preprocess: couldn't find program "minctracc" > > fmr_preprocess: couldn't find program "param2xfm" > > > > Usage: fmr_preprocess input.mnc [input2.mnc ...] -options > > > > It seems some options of fmr_preprocess doesn't work well due to lack of > > these programs. Such as -blur and -align. > > > > How can i fix it? > > > > Thank you for helping us. > > > > Cheers, > > Sheng > > > > On 7/4/06, Andrew Janke wrote: > > > > > > > So you mean the distribution you are using is Ubuntu? You can run > all > > > > software ,including fmr_preprocess, successfully on Ubuntu? what is > it > > > > version? the latest one Dapper(6.10)? > > > > > > At the BIC we primarilly use and support Debian sarge. (with a 2.6.8 > > > kernel or 2.4.27). Others in the BIC are using Ubuntu but I don't > > > know which version. > > > > > > > > > a > > > _______________________________________________ > > > MINC-users at bic.mni.mcgill.ca > > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > > > > > > > > > -- > > EE is no more about circuits than CS is about computers. > > > > Home Page: http://forrest.bao.googlepages.com > > Blog: http://blog.360.yahoo.com/forrestbao > > http://forrestbao.spaces.msn.com > > MSN: bsh1984 at hotmail.com (IM only) > > Gmail: forrest.bao at gmail.com (Email also) > > Yahoo! ID: forrestbao at yahoo.com.cn (Email also) > > AIM/ICQ: forrestbao > > Skype: forrestbao > > _______________________________________________ > > MINC-users at bic.mni.mcgill.ca > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > > > -- > Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) > Canada->Montreal Cell: +1 (514) 924 2012 > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- EE is no more about circuits than CS is about computers. Home Page: http://forrest.bao.googlepages.com Blog: http://blog.360.yahoo.com/forrestbao http://forrestbao.spaces.msn.com MSN: bsh1984 at hotmail.com (IM only) Gmail: forrest.bao at gmail.com (Email also) Yahoo! ID: forrestbao at yahoo.com.cn (Email also) AIM/ICQ: forrestbao Skype: forrestbao From forrest.bao at gmail.com Sat Jul 15 22:29:59 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Sun, 16 Jul 2006 10:29:59 +0800 Subject: [MINC-users] where does these programs contain? In-Reply-To: <5D3BE852-E2AF-4982-9E58-E8C0205AF420@bic.mni.mcgill.ca> References: <889df5f00607100248i6ff6adb4sdd4ddf95f9fdee9c@mail.gmail.com> <5D3BE852-E2AF-4982-9E58-E8C0205AF420@bic.mni.mcgill.ca> Message-ID: <889df5f00607151929j1368fb54x14f28f323b6ef1bf@mail.gmail.com> yes, I have handle everything. And they all work well now. Thank you very much. On 7/10/06, Jonathan Harlap wrote: > > Hi Sheng, > > These programs are part of the mni_autoreg package. > > Cheers, > J > > On 10-Jul-06, at 5:48 AM, Sheng Bao wrote: > > > Hi guys, I have installed MINC and BICPL but when I tries to run > > fmr_preprocess, I found some programs can't be found. I have > > specified the > > location to install programs by the --prefix arguement. And I can't > > find > > them at /usr/local/mni/bin > > > > So, can anyone tell me which packages contain these programs? These > > programs > > are : mincblur, alignnear, autocrop, minctracc, param2xfm > > > > root at orioleQ:~/fmri/bicpl-1.4.3# fmr_preprocess > > fmr_preprocess: couldn't find program "mincblur" > > fmr_preprocess: couldn't find program "alignlinear" > > fmr_preprocess: couldn't find program "autocrop" > > fmr_preprocess: couldn't find program "minctracc" > > fmr_preprocess: couldn't find program "param2xfm" > > > > Usage: fmr_preprocess input.mnc [input2.mnc ...] -options > > > > Thank you for helping me. > > > > Cheers, > > Sheng > > > > -- > > EE is no more about circuits than CS is about computers. > > > > Home Page: http://forrest.bao.googlepages.com > > Blog: http://blog.360.yahoo.com/forrestbao > > http://forrestbao.spaces.msn.com > > MSN: bsh1984 at hotmail.com (IM only) > > Gmail: forrest.bao at gmail.com (Email also) > > Yahoo! ID: forrestbao at yahoo.com.cn (Email also) > > AIM/ICQ: forrestbao > > Skype: forrestbao > > _______________________________________________ > > MINC-users at bic.mni.mcgill.ca > > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- EE is no more about circuits than CS is about computers. Home Page: http://forrest.bao.googlepages.com Blog: http://blog.360.yahoo.com/forrestbao http://forrestbao.spaces.msn.com MSN: bsh1984 at hotmail.com (IM only) Gmail: forrest.bao at gmail.com (Email also) Yahoo! ID: forrestbao at yahoo.com.cn (Email also) AIM/ICQ: forrestbao Skype: forrestbao From forrest.bao at gmail.com Tue Jul 18 03:30:32 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Tue, 18 Jul 2006 15:30:32 +0800 Subject: [MINC-users] I plan to write a new article about MINC at Linux Focus Message-ID: <889df5f00607180030n30354095p9b9e6ce9b93d47f9@mail.gmail.com> Hi guys( the people of LinuxFocus and MINC users), I wanna to write a new article about fMRI and Brain Mapping research toolchain on Linux. Since few weeks ago, I joined a cooperation research of Montreal Neural Institute, McGill University, Canada and Nanjing Jing-Ling Hospital, China, I began to touch this fields. In that research, I helped maintaning Linux work stations in Nanjing, coz no one in the Dept. of Med. Img. is familiar with Linux. And I found many workable tools for this reseach on Linux. I plan to write a series of articles. On the first article, I will introduce the MINC developed by BIC, MNI, McGill university. This is the tool that we are using now. Contents of this article shall be installation instructions, brief usage, and other issues. I wanna write this article coz a professor from Chinese Academy of Science, who has cooperation with us now, asked me to give him an instruction of how to install fmr_preprocess. Here is a snapshot of Register program in MINC: http://picasaweb.google.com/forrest.bao/BrainImaging/photo#4949571112904753170 Then I plan to write other articles introducing other tools on Linux for this research, both commercial and open-source ones. Above is my brief plan. I will start my writing after I finish revising an on-going paper. Cheers, Sheng -- Sheng Bao, research assistant Dept. of Medical Image, Nanjing Jingling Hospital Clinical School, Medicine College, Nanjing University Home Page: http://forrest.bao.googlepages.com MSN: bsh1984 at hotmail.com Yahoo! ID: forrestbao at yahoo.com.cn Gmail: forrest.bao at gmail.com AIM/ICQ: forrestbao Skype: forrestbao [image: Get Firefox!] From hehuiguang at gmail.com Tue Jul 18 04:09:34 2006 From: hehuiguang at gmail.com (Huiguang He) Date: Tue, 18 Jul 2006 16:09:34 +0800 Subject: [MINC-users] about mritotal Message-ID: <2c8c005e0607180109u2a56aaeaxa8d5a8daf578ff48@mail.gmail.com> Hi, Dear Minc Expert: I have a question about the command mritotal. As we know, mritotal is the command that convert the mridata into Talrache coordinate. But does it require that the input data have the same orientation as the template? I found that the template is sagital orientation, but our fMRI scanning is coronary orientation, should I re-orientate my data to the same as template? Or mritotal can handle this issue and give us a correct transform matrix? Thanks a lot! Huiguang Huiguang He, Ph.D Chinese Academy of Sciences From a.janke at gmail.com Tue Jul 18 08:59:04 2006 From: a.janke at gmail.com (Andrew Janke) Date: Tue, 18 Jul 2006 08:59:04 -0400 Subject: [MINC-users] about mritotal In-Reply-To: <2c8c005e0607180109u2a56aaeaxa8d5a8daf578ff48@mail.gmail.com> References: <2c8c005e0607180109u2a56aaeaxa8d5a8daf578ff48@mail.gmail.com> Message-ID: Huigang, Yes, the data will have to be correctly oriented before mritotal will work correctly. Remember that in MINC data is _always_ supposed to be correctly oriented. This means that if you have done a coronal or sagittal acquisition then all that will be different in your file is the dimension ordering. If you are converting from dicom using dcm2mnc you should have no problems with this. Andrew On 7/18/06, Huiguang He wrote: > Hi, Dear Minc Expert: > I have a question about the command mritotal. > As we know, mritotal is the command that convert the mridata into > Talrache coordinate. But does it require that the input data have the same > orientation as the template? > I found that the template is sagital orientation, but our fMRI > scanning is coronary orientation, should I re-orientate my data to the same > as template? Or mritotal can handle this issue and give us a correct > transform matrix? > Thanks a lot! > Huiguang > > Huiguang He, Ph.D > Chinese Academy of Sciences > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From a.janke at gmail.com Tue Jul 18 10:56:14 2006 From: a.janke at gmail.com (Andrew Janke) Date: Tue, 18 Jul 2006 10:56:14 -0400 Subject: [MINC-users] I plan to write a new article about MINC at Linux Focus In-Reply-To: <889df5f00607180030n30354095p9b9e6ce9b93d47f9@mail.gmail.com> References: <889df5f00607180030n30354095p9b9e6ce9b93d47f9@mail.gmail.com> Message-ID: Hi Sheng, This sounds like a good idea to me. I would appreciate it if you could forward a copy of what you plan on publishing here (or to me) so that we/I can add any extra things that you may not be aware of yet. It might also be an idea to develop this article(s) on the bic wiki (wiki.bic.mni.mcgill.ca) then we will have a local permanent record of what you have done. Feel free to make your own page of instructions/etc on there as per usual wiki style. Good luck. Andrew On 7/18/06, Sheng Bao wrote: > Hi guys( the people of LinuxFocus and MINC users), > I wanna to write a new article about fMRI and Brain Mapping research > toolchain on Linux. Since few weeks ago, I joined a cooperation research of > Montreal Neural Institute, McGill University, Canada and Nanjing Jing-Ling > Hospital, China, I began to touch this fields. In that research, I helped > maintaning Linux work stations in Nanjing, coz no one in the Dept. of Med. > Img. is familiar with Linux. And I found many workable tools for this > reseach on Linux. > > I plan to write a series of articles. On the first article, I will > introduce the MINC developed by BIC, MNI, McGill university. This is the > tool that we are using now. Contents of this article shall be installation > instructions, brief usage, and other issues. I wanna write this article coz > a professor from Chinese Academy of Science, who has cooperation with us > now, asked me to give him an instruction of how to install fmr_preprocess. > > Here is a snapshot of Register program in MINC: > http://picasaweb.google.com/forrest.bao/BrainImaging/photo#4949571112904753170 > > Then I plan to write other articles introducing other tools on Linux for > this research, both commercial and open-source ones. > > Above is my brief plan. I will start my writing after I finish revising an > on-going paper. > > Cheers, > Sheng > > > -- > Sheng Bao, research assistant > Dept. of Medical Image, Nanjing Jingling Hospital > Clinical School, Medicine College, Nanjing University > Home Page: http://forrest.bao.googlepages.com > MSN: bsh1984 at hotmail.com > Yahoo! ID: forrestbao at yahoo.com.cn > Gmail: forrest.bao at gmail.com > AIM/ICQ: forrestbao > Skype: forrestbao > [image: Get Firefox!] > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From forrest.bao at gmail.com Wed Jul 19 03:08:12 2006 From: forrest.bao at gmail.com (Sheng Bao) Date: Wed, 19 Jul 2006 15:08:12 +0800 Subject: [MINC-users] the distribution variancies problem is boring Message-ID: <889df5f00607190008v29490c25m473971d88b81d288@mail.gmail.com> TO ZQ Zhang, As you known, there are many different distributions of Linux, Fedora Core, Debian, SuSE, Ubuntu, etc. They use the same kernel, the Linux. But they have different libraries which will be called by other programs. Only the kernel is not enough to run a computer. Linux is wholly open. So different software use different libraries. Different distributions have different libraries. Thus, the software that I can install on one distribution may not work on another distribution. Here is a famous explanation to the dark side of open source strategy: http://zone.ni.com/devzone/conceptd.nsf/2d17d611efb58b22862567a9006ffe76/9d1bcf97a0aaedba8625685e00753206?OpenDocument&node=145000_US#3 It is sometimes boring with software installation problem due to the open of Linux. In our case, maybe we can run MINC on one distribution but due to other reason, we can't run freesurfer on that distribution. And when we have immigrated to another distribution, we found that MIPAV doesn't work. I chose to use Fedora Core at the beginning was to avoid the lack of libraries. But now, I found many medical imaging software are developed on the Debian platform. And respecting to the convenient package management theme of Ubuntu, I think we may better test whether all software that we may use can run on Ubuntu. Anyway, if they can work well on SuSE or Fedora Core 4, it is our best luck. Now I think Ubuntu is appropriate to install MINC coz I don't wanna install glut libs on Fedora Core 4 manually but via apt-get. PS: I think MINC should make a list of recommended distribution. At least now, I have tested SuSE 10, Fedora Core 4 and Ubuntu 6.10 -- Sheng Bao, research assistant Dept. of Medical Image, Nanjing Jingling Hospital Clinical School, Medicine College, Nanjing University Home Page: http://forrest.bao.googlepages.com MSN: bsh1984 at hotmail.com Yahoo! ID: forrestbao at yahoo.com.cn Gmail: forrest.bao at gmail.com AIM/ICQ: forrestbao Skype: forrestbao From a.janke at gmail.com Wed Jul 19 09:22:57 2006 From: a.janke at gmail.com (Andrew Janke) Date: Wed, 19 Jul 2006 09:22:57 -0400 Subject: [MINC-users] transformation of GE to MINC In-Reply-To: <44B78A81.000106.26615@m64.126.com> References: <44B78A81.000106.26615@m64.126.com> Message-ID: > I encountered a problem when I transformate GE to MINC. we acquired data with GE-signa 1.5 Tesla scanner, and wanted ultimate data with MINC format. for this sake. we tested 3 ways: > 1. transformate GE data to DICOM, then to MINC. At the ADW4.0 workstation of GE, exported the data of GE format to ADW to finish DICOM transformation. but there is a trouble that one folder can only hold 1000 files in GE equipment, 3000 files of one run in our experiment were divided into 3 folder, they can't be accepted by the programme of dcm2mnc(BIC MNI McGill). we tried to rename 3000 fiels, and packed them into one folder, the dcm2mnc can't read it at all. then we tested the second way: > 2. transformated DICOM to analyze. after renaming and packing 3000 dicom files, I finshed analyze transormation using [MRIcro/Import/convert foreign to analyze]. then transformated it to MINC, but next step(Register program at BIC.MNI.McGILL) go wrong, 2 images can't be registered. > 3. transformate GE to MINC, after renaming and packing 3000 GE files to analyze(*.img&*.hdr) I used program ge5_to_minc to finish it. but ge5_to_minc can't run at all, it displayed: "1024 error". Zhang, I suspect the least problematic method will be #1. However when you say that dc2mnc cant read it what was the error? If you have all the dicom files in one folder you can call dcm2mnc as such: $ dcm2mnc input_dir output_dir I suggest you also make sure that you are using a recent version of dcm2mnc, the one included in minc 1.4 should be enough. -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From sean at rogue-research.com Wed Jul 19 12:16:03 2006 From: sean at rogue-research.com (Sean McBride) Date: Wed, 19 Jul 2006 12:16:03 -0400 Subject: [MINC-users] Using MINC on Intel Macs In-Reply-To: <44B3AB29.6090708@bic.mni.mcgill.ca> References: <997DCA7B-DF04-46A7-9D38-1C74D3986D21@loni.ucla.edu> <44B3AB29.6090708@bic.mni.mcgill.ca> Message-ID: <20060719161603.14786@smtp1.sympatico.ca> On 2006-07-11 09:44, Jason Lerch said: >A full installer (370Mb) can be found at: > >http://www.bic.mni.mcgill.ca/users/jason/mactel/ > >This will install into /usr/local/minc2 (well, it's supposed to, so let >me know if it decides to put the software somewhere else). Everything is >compiled against a modified version of minc 2.0.09. > >Let me know of any bugs anyone finds. And if you do, do so quickly - I >too will lose easy access to an intel mac by the end of the week. Hi Jason, Thanks for making these available. But could you update the various webpages? The main MINC page at has links to sources and a Windows binary installation, but no Mac links at all. This page: is 2 years old, and has only minc 1.3. Then there are these 2 links: but they show just file listings. Basically, I find it all rather confusing. :) Thanks, -- ____________________________________________________________ Sean McBride, B. Eng sean at rogue-research.com Rogue Research www.rogue-research.com Mac Software Developer Montr?al, Qu?bec, Canada From sorench at gmail.com Thu Jul 20 00:18:33 2006 From: sorench at gmail.com (Soren Christensen) Date: Thu, 20 Jul 2006 14:18:33 +1000 Subject: [MINC-users] mnc2nii spatial issue Message-ID: Hi, I am having a problem with mnc2nii and nii2mnc. When converting a file from mnc to nifti and then back to minc the dimension startpoints can vary. See mincinfo below. Register confirms that the actual voxel values at synced coordinates are slightly different. Is this a bug or am I missing something. I added the header info described in http://www.bic.mni.mcgill.ca/pipermail/minc-users/2005-October/000857.html just in case, but that issue seems to be resolved. Cheers Soren mincinfo labt.vol.mnc labbott.vol_new.mnc file: labt.vol.mnc image: signed__ short 0 to 397 image dimensions: yspace zspace xspace dimension name length step start -------------- ------ ---- ----- yspace 120 1.5 -384 zspace 512 0.46875 -28.125 xspace 512 -0.46875 120 file: labbott.vol_new.mnc image: signed__ float 0 to 397 image dimensions: zspace yspace xspace dimension name length step start -------------- ------ ---- ----- zspace 512 0.46875 -28.125 yspace 120 1.5 -384 xspace 512 0.46875 -119.531 From sorench at gmail.com Thu Jul 20 01:09:44 2006 From: sorench at gmail.com (Soren Christensen) Date: Thu, 20 Jul 2006 15:09:44 +1000 Subject: [MINC-users] nii2mnc Message-ID: Hi guys, Sorry for the previous post. Realized that the startpoint was different as the axis was flipped (inverted sign of step size). The value was different due to the different datatype (float vs signed short). Sorry again, Soren From Michel.Audette at medizin.uni-leipzig.de Sat Jul 22 03:20:00 2006 From: Michel.Audette at medizin.uni-leipzig.de (Audette, Michel) Date: Sat, 22 Jul 2006 09:20:00 +0200 Subject: [MINC-users] dcm2mnc Skipping file ... which is not in the expected format References: <160E3DD4FB702C4CB860C65186686E69757A26@MRZS152229.medizin.uni-leipzig.de><160E3DD4FB702C4CB860C65186686E6965874D@MRZS152229.medizin.uni-leipzig.de><371D07BA-3CE2-4F4F-B842-B5A0E06E7C6D@bic.mni.mcgill.ca> <160E3DD4FB702C4CB860C65186686E69AC3A57@MRZS152229.medizin.uni-leipzig.de> Message-ID: <160E3DD4FB702C4CB860C65186686E69AC3A58@MRZS152229.medizin.uni-leipzig.de> Hi everyone, I have a strange problem with dcm2mnc, whereby I get... Skipping file ..., which is not in the expected format, and no minc volume. Recently, I reinstalled my minc2 software, because I mistakenly overwrote it with minc1 stuff (I had installed old minc stuff to interact with software from INRIA, which features some minc and netcdf calls but does not yet support minc2). I used to get a lot of complaining, WARNING: Can't find Philips slice index, but after all the complaints, what I had managed to muddle through and produce a minc volume ultimately. Now, just the Skipping file stuff, and no minc volume. Can anyone suggest how to debug this? Best regards, Michel Michel Audette, Ph.D. Innovation Center Computer Assisted Surgery (ICCAS) Philipp-Rosenthal-Strasse 55 Leipzig, Germany Phone: ++49 (0) 341 / 97 - 1 20 13 Fax: ++49 (0) 341 / 97 - 1 20 09 From abraham.gedamu at gmail.com Tue Jul 25 15:01:08 2006 From: abraham.gedamu at gmail.com (Abraham Gedamu) Date: Tue, 25 Jul 2006 15:01:08 -0400 Subject: [MINC-users] Registration using only translation and scale in minc Message-ID: Hello, I would like to register one image to another using only translation and scale. I have tried using minctracc but there are no options that allow the use of only translation and scale. I could get all the parameters and set the ones I don't want to zero, but then the values for scale and translation will not be the correct ones since the paramters seem to depend on one another. Any advice in this are would be greatly appreciated. Thanks Abe From a.janke at gmail.com Wed Jul 26 10:18:54 2006 From: a.janke at gmail.com (Andrew Janke) Date: Wed, 26 Jul 2006 10:18:54 -0400 Subject: [MINC-users] Registration using only translation and scale in minc In-Reply-To: References: Message-ID: Abe, You need to set the weight parameters for optimisation in minctracc. So for what you want: minctracc -lsq9 -w_rotations 0 0 0 .... should do what you want. a On 7/25/06, Abraham Gedamu wrote: > Hello, > > I would like to register one image to another using only translation and > scale. I have tried using minctracc but there are no options that allow the > use of only translation and scale. I could get all the parameters and set > the ones I don't want to zero, but then the values for scale and translation > will not be the correct ones since the paramters seem to depend on one > another. Any advice in this are would be greatly appreciated. > > Thanks > > Abe > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From pgravel at bic.mni.mcgill.ca Wed Jul 26 12:57:39 2006 From: pgravel at bic.mni.mcgill.ca (Paul GRAVEL) Date: Wed, 26 Jul 2006 12:57:39 -0400 Subject: [MINC-users] mincresample changes voxel values Message-ID: Dear All, I am combining manually_modified_labels using Display: Display mri_stx_nuc.mnc -label label1.mnc -label label2.mnc ... (All the label files are in the icbm_template_2.00mm.mnc format/dimensions/origin.) Then when I save the combined labels in Display, it does put the labels in a different format than the icbm; which I guess is expected. But when I use mincresample on the combined labels to bring them back to the icbm_template_2.00mm.mnc format, some of the original label numbers are changed, e.g. Original right_hippocampus_voxels=36, but resampled right_hippocampus will have some voxels=36 and some voxels=37. My question is why is that, and/or is there a work-around ? I did use: mincresample combined_labels_Display_format.mnc combined_labels_icbm_2mm.mnc -like /data/avgbrain/avgbrain1/brain/images/icbm_template_2.00mm.mnc OR mincresample -nearest_neighbour combined_labels_Display_format.mnc combined_labels_icbm_2mm.mnc -like /data/avgbrain/avgbrain1/brain/images/icbm_template_2.00mm.mnc But same outcome... Thanks in advance. Best Regards, Paul From jason at bic.mni.mcgill.ca Wed Jul 26 13:04:06 2006 From: jason at bic.mni.mcgill.ca (Jason Lerch) Date: Wed, 26 Jul 2006 13:04:06 -0400 Subject: [MINC-users] mincresample changes voxel values In-Reply-To: References: Message-ID: <44C7A086.9010406@bic.mni.mcgill.ca> Quick word of caution: are you reading these values using Display or register? They do their own internal conversion - to make sure this problem is really a problem, try binarizing a label (use mincmath -const2 35.5 36.5 -seg labels.mnc one-label.mnc) and then loading it in some viewing program. Hope this helps - if not, see if the command resample_labels gives you more satisfaction. Jason Paul GRAVEL wrote: >Dear All, > >I am combining manually_modified_labels using Display: > >Display mri_stx_nuc.mnc -label label1.mnc -label label2.mnc ... > >(All the label files are in the icbm_template_2.00mm.mnc >format/dimensions/origin.) > >Then when I save the combined labels in Display, it does put the labels in >a different format than the icbm; which I guess is expected. > >But when I use mincresample on the combined labels to bring them back to >the icbm_template_2.00mm.mnc format, some of the original label numbers >are changed, e.g. Original right_hippocampus_voxels=36, but resampled >right_hippocampus will have some voxels=36 and some voxels=37. >My question is why is that, and/or is there a work-around ? > >I did use: >mincresample combined_labels_Display_format.mnc combined_labels_icbm_2mm.mnc -like /data/avgbrain/avgbrain1/brain/images/icbm_template_2.00mm.mnc >OR >mincresample -nearest_neighbour combined_labels_Display_format.mnc combined_labels_icbm_2mm.mnc -like /data/avgbrain/avgbrain1/brain/images/icbm_template_2.00mm.mnc > >But same outcome... > >Thanks in advance. > >Best Regards, > >Paul > > > > > >_______________________________________________ >MINC-users at bic.mni.mcgill.ca >http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > From abraham.gedamu at gmail.com Wed Jul 26 14:28:58 2006 From: abraham.gedamu at gmail.com (Abraham Gedamu) Date: Wed, 26 Jul 2006 14:28:58 -0400 Subject: [MINC-users] Registration using only translation and scale in minc In-Reply-To: References: Message-ID: I tried minctracc -lsq9 w_rotations 0 0 0 source.mnc target.mnc out.xfmseveral times with different source files and target files, however when I run xfm2param on the out.xfm file, my rotations are not 0. Does anyone no why this might occur? Thanks, Abe On 7/25/06, Abraham Gedamu wrote: > > Hello, > > I would like to register one image to another using only translation and > scale. I have tried using minctracc but there are no options that allow the > use of only translation and scale. I could get all the parameters and set > the ones I don't want to zero, but then the values for scale and translation > will not be the correct ones since the paramters seem to depend on one > another. Any advice in this are would be greatly appreciated. > > Thanks > > Abe > From atsuko at bic.mni.mcgill.ca Wed Jul 26 16:30:41 2006 From: atsuko at bic.mni.mcgill.ca (Atsuko Nagano) Date: Wed, 26 Jul 2006 16:30:41 -0400 Subject: [MINC-users] ROI template for DLPFC Message-ID: <16761c140607261330w208475efqfbb1a249cabd702a@mail.gmail.com> Hello all, Does anybody have ROI template for DLPFC. I am interested in the borders of the BA46. BA9/46d. and BA9/46v. Best, Atsuko From pgravel at bic.mni.mcgill.ca Wed Jul 26 20:56:07 2006 From: pgravel at bic.mni.mcgill.ca (Paul GRAVEL) Date: Wed, 26 Jul 2006 20:56:07 -0400 Subject: [MINC-users] mincresample changes voxel values In-Reply-To: <44C7A086.9010406@bic.mni.mcgill.ca> Message-ID: Thanks Jason! The program reading these values is do_GTM_correction (program to correct for partial volume effect). The new label 37 was found in the output of this program. I then used Display, which confirmed that there was indeed a new label 37; about half of the right hippocampus was changed to 37, while the rest remained to 36. I did try the mincmath command, and it seems to work, i.e. it changed the whole (and not half of) right hippocampus to label 1. But I am now using: mincreshape -dimorder 'zspace yspace xspace' -dimsize 'zspace=91' -dimsize 'yspace=109' -dimsize 'xspace=91' -start '-1 -1 4' combined_labels_Display_format.mnc combined_labels_icbm_2mm.mnc and it seems to do the trick. Thanks again! Paul On Wed, 26 Jul 2006, Jason Lerch wrote: > Quick word of caution: are you reading these values using Display or > register? They do their own internal conversion - to make sure this > problem is really a problem, try binarizing a label (use mincmath > -const2 35.5 36.5 -seg labels.mnc one-label.mnc) and then loading it in > some viewing program. > > Hope this helps - if not, see if the command resample_labels gives you > more satisfaction. > > Jason > > Paul GRAVEL wrote: > > >Dear All, > > > >I am combining manually_modified_labels using Display: > > > >Display mri_stx_nuc.mnc -label label1.mnc -label label2.mnc ... > > > >(All the label files are in the icbm_template_2.00mm.mnc > >format/dimensions/origin.) > > > >Then when I save the combined labels in Display, it does put the labels in > >a different format than the icbm; which I guess is expected. > > > >But when I use mincresample on the combined labels to bring them back to > >the icbm_template_2.00mm.mnc format, some of the original label numbers > >are changed, e.g. Original right_hippocampus_voxels=36, but resampled > >right_hippocampus will have some voxels=36 and some voxels=37. > >My question is why is that, and/or is there a work-around ? > > > >I did use: > >mincresample combined_labels_Display_format.mnc combined_labels_icbm_2mm.mnc -like /data/avgbrain/avgbrain1/brain/images/icbm_template_2.00mm.mnc > >OR > >mincresample -nearest_neighbour combined_labels_Display_format.mnc combined_labels_icbm_2mm.mnc -like /data/avgbrain/avgbrain1/brain/images/icbm_template_2.00mm.mnc > > > >But same outcome... > > > >Thanks in advance. > > > >Best Regards, > > > >Paul > > > > > > > > > > > >_______________________________________________ > >MINC-users at bic.mni.mcgill.ca > >http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > > > > > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > From a.janke at gmail.com Wed Jul 26 23:37:12 2006 From: a.janke at gmail.com (Andrew Janke) Date: Wed, 26 Jul 2006 23:37:12 -0400 Subject: [MINC-users] ROI template for DLPFC In-Reply-To: <16761c140607261330w208475efqfbb1a249cabd702a@mail.gmail.com> References: <16761c140607261330w208475efqfbb1a249cabd702a@mail.gmail.com> Message-ID: > Does anybody have ROI template for DLPFC. > I am interested in the borders of the BA46. BA9/46d. and BA9/46v. Not that I know of. However be very careful mixing "mni-space", talairach and brodman areas. The mni-models (icbm, average 305, etc) are not in exact talairach space due to various constraints, thus we call this "mni-space". The brodman areas you talk of that I presume you will be mapping from a talairach atlas will not align. (perhaps within 5mm at some points but not at all in others). You'd be best to identify your own ROI on the ICBM average (or your own average of your data) using Display and then go from there. -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From Michel.Audette at medizin.uni-leipzig.de Thu Jul 27 09:46:48 2006 From: Michel.Audette at medizin.uni-leipzig.de (Audette, Michel) Date: Thu, 27 Jul 2006 15:46:48 +0200 Subject: [MINC-users] ensuring that mincreshape produces positive axes Message-ID: <160E3DD4FB702C4CB860C65186686E69AC3A73@MRZS152229.medizin.uni-leipzig.de> Hi everyone, I'm playing with minc volumes in Vtk, and I found that Marching Cubes seems to crash with negative axes. To resolve this, I tried mincreshape, only to get 2 out 3 axes become positive. maudette at icaw164201:~/research/source/work> mincreshape -clobber +direction becken_cancellous_crop3_blur.mnc becken_cancellous_crop3_blur_plus.mnc Copying chunks:..................................................................................................................................................................................................................................................................................................Done. maudette at icaw164201:~/research/source/work> mincinfo becken_cancellous_crop3_blur_plus.mnc file: becken_cancellous_crop3_blur_plus.mnc image: unsigned byte 0 to 255 image dimensions: zspace yspace xspace dimension name length step start -------------- ------ ---- ----- zspace 290 -0.732422 -1423.84 yspace 209 0.732422 -217.525 xspace 191 0.732422 -33.7188 Is there a trick to achieving this, or do I have to use mincresample and re-specify start points? Cheers, Michel From a.janke at gmail.com Thu Jul 27 10:13:01 2006 From: a.janke at gmail.com (Andrew Janke) Date: Thu, 27 Jul 2006 10:13:01 -0400 Subject: [MINC-users] ensuring that mincreshape produces positive axes In-Reply-To: <160E3DD4FB702C4CB860C65186686E69AC3A73@MRZS152229.medizin.uni-leipzig.de> References: <160E3DD4FB702C4CB860C65186686E69AC3A73@MRZS152229.medizin.uni-leipzig.de> Message-ID: Take a look at man mincreshape, there is a "extra thing" that you have to do to force mincreshape to flip the slice dimension. This will always work (and also make everything zyx ordered. mincreshape \ +direction \ -dimorder zspace,yspace,xspace \ -dimsize xspace=-1 \ -dimsize yspace=-1 \ -dimsize zspace=-1 \ in.mnc out.mnc a On 7/27/06, Audette, Michel wrote: > Hi everyone, > > I'm playing with minc volumes in Vtk, and I found that Marching Cubes seems to crash with negative axes. To resolve this, I tried mincreshape, only to get 2 out 3 axes become positive. > > maudette at icaw164201:~/research/source/work> mincreshape -clobber +direction becken_cancellous_crop3_blur.mnc becken_cancellous_crop3_blur_plus.mnc > Copying chunks:..................................................................................................................................................................................................................................................................................................Done. > maudette at icaw164201:~/research/source/work> mincinfo becken_cancellous_crop3_blur_plus.mnc > file: becken_cancellous_crop3_blur_plus.mnc > image: unsigned byte 0 to 255 > image dimensions: zspace yspace xspace > dimension name length step start > -------------- ------ ---- ----- > zspace 290 -0.732422 -1423.84 > yspace 209 0.732422 -217.525 > xspace 191 0.732422 -33.7188 > > Is there a trick to achieving this, or do I have to use mincresample and re-specify start points? > > Cheers, > > Michel > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 From Michel.Audette at medizin.uni-leipzig.de Thu Jul 27 10:25:56 2006 From: Michel.Audette at medizin.uni-leipzig.de (Audette, Michel) Date: Thu, 27 Jul 2006 16:25:56 +0200 Subject: [MINC-users] ensuring that mincreshape produces positive axes References: <160E3DD4FB702C4CB860C65186686E69AC3A73@MRZS152229.medizin.uni-leipzig.de> Message-ID: <160E3DD4FB702C4CB860C65186686E69AC3A74@MRZS152229.medizin.uni-leipzig.de> Hi Andrew, thanks for both this response and the previous one on mincmorph, which I failed to answer. You're the cornerstone of minc-users. I'll give it a try, although I need x-y-z ordering. Cheers, Michel -----Original Message----- From: minc-users-bounces at bic.mni.mcgill.ca on behalf of Andrew Janke Sent: Thu 7/27/2006 4:13 PM To: MINC users mailing list Subject: Re: [MINC-users] ensuring that mincreshape produces positive axes Take a look at man mincreshape, there is a "extra thing" that you have to do to force mincreshape to flip the slice dimension. This will always work (and also make everything zyx ordered. mincreshape \ +direction \ -dimorder zspace,yspace,xspace \ -dimsize xspace=-1 \ -dimsize yspace=-1 \ -dimsize zspace=-1 \ in.mnc out.mnc a On 7/27/06, Audette, Michel wrote: > Hi everyone, > > I'm playing with minc volumes in Vtk, and I found that Marching Cubes seems to crash with negative axes. To resolve this, I tried mincreshape, only to get 2 out 3 axes become positive. > > maudette at icaw164201:~/research/source/work> mincreshape -clobber +direction becken_cancellous_crop3_blur.mnc becken_cancellous_crop3_blur_plus.mnc > Copying chunks:..................................................................................................................................................................................................................................................................................................Done. > maudette at icaw164201:~/research/source/work> mincinfo becken_cancellous_crop3_blur_plus.mnc > file: becken_cancellous_crop3_blur_plus.mnc > image: unsigned byte 0 to 255 > image dimensions: zspace yspace xspace > dimension name length step start > -------------- ------ ---- ----- > zspace 290 -0.732422 -1423.84 > yspace 209 0.732422 -217.525 > xspace 191 0.732422 -33.7188 > > Is there a trick to achieving this, or do I have to use mincresample and re-specify start points? > > Cheers, > > Michel > > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 _______________________________________________ MINC-users at bic.mni.mcgill.ca http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users From tguo at imaging.robarts.ca Thu Jul 27 10:32:34 2006 From: tguo at imaging.robarts.ca ((Jessie) Ting Guo) Date: Thu, 27 Jul 2006 10:32:34 -0400 Subject: [MINC-users] Question with mincedit Message-ID: <005c01c6b189$803af060$c32914c6@irus.robarts.ca> Hi, I'm trying to edit the head file of a minc image using mincedit. However I got the memory shortage problem no matter which editor I was using. The RAM on my desktop is about 2G. Could anyone tell me how to fix this? Thank you very much! Jessie From a.janke at gmail.com Thu Jul 27 10:36:55 2006 From: a.janke at gmail.com (Andrew Janke) Date: Thu, 27 Jul 2006 10:36:55 -0400 Subject: [MINC-users] Registration using only translation and scale in minc In-Reply-To: References: Message-ID: On 7/26/06, Abraham Gedamu wrote: > I tried minctracc -lsq9 w_rotations 0 0 0 source.mnc target.mnc > out.xfmseveral times with different source files and target files, > however when I > run xfm2param on the out.xfm file, my rotations are not 0. What are the rotations? very close to 0? If so then I suspect you are being bitten by the fact that there are a number of ways to describe the transformation in an affine matrix with a set of translations, scales and rotations. However if you are only trying to recover translation and scale this still should not happen. Copy in what minctracc and param2xfm return (And what you entered) and we'll go from there. a From a.janke at gmail.com Thu Jul 27 10:42:01 2006 From: a.janke at gmail.com (Andrew Janke) Date: Thu, 27 Jul 2006 10:42:01 -0400 Subject: [MINC-users] dcm2mnc Skipping file ... which is not in the expected format In-Reply-To: <160E3DD4FB702C4CB860C65186686E69AC3A58@MRZS152229.medizin.uni-leipzig.de> References: <160E3DD4FB702C4CB860C65186686E69757A26@MRZS152229.medizin.uni-leipzig.de> <160E3DD4FB702C4CB860C65186686E6965874D@MRZS152229.medizin.uni-leipzig.de> <371D07BA-3CE2-4F4F-B842-B5A0E06E7C6D@bic.mni.mcgill.ca> <160E3DD4FB702C4CB860C65186686E69AC3A57@MRZS152229.medizin.uni-leipzig.de> <160E3DD4FB702C4CB860C65186686E69AC3A58@MRZS152229.medizin.uni-leipzig.de> Message-ID: > I have a strange problem with dcm2mnc, whereby I get... > > Skipping file ..., which is not in the expected format, and no minc volume. > > Recently, I reinstalled my minc2 software, because I mistakenly overwrote it with minc1 stuff (I had installed old minc stuff to interact with software from INRIA, which features some minc and netcdf calls but does not yet support minc2). > > I used to get a lot of complaining, WARNING: Can't find Philips slice index, > but after all the complaints, what I had managed to muddle through and produce a minc volume ultimately. > > Now, just the Skipping file stuff, and no minc volume. > > Can anyone suggest how to debug this? My only suggestion is to add the -debug flag and then send the (condensed/pertinent) output here and then wait until Bert Vincent fixes it.. :) dcm2mnc is still to my knowledge his baby although a few others have taken some interest in it.. a From a.janke at gmail.com Thu Jul 27 10:42:28 2006 From: a.janke at gmail.com (Andrew Janke) Date: Thu, 27 Jul 2006 10:42:28 -0400 Subject: [MINC-users] dcm2mnc Skipping file ... which is not in the expected format In-Reply-To: References: <160E3DD4FB702C4CB860C65186686E69757A26@MRZS152229.medizin.uni-leipzig.de> <160E3DD4FB702C4CB860C65186686E6965874D@MRZS152229.medizin.uni-leipzig.de> <371D07BA-3CE2-4F4F-B842-B5A0E06E7C6D@bic.mni.mcgill.ca> <160E3DD4FB702C4CB860C65186686E69AC3A57@MRZS152229.medizin.uni-leipzig.de> <160E3DD4FB702C4CB860C65186686E69AC3A58@MRZS152229.medizin.uni-leipzig.de> Message-ID: > My only suggestion is to add the -debug flag and then send the > (condensed/pertinent) output here oops, make that the -verbose flag. a From abraham.gedamu at gmail.com Thu Jul 27 10:51:07 2006 From: abraham.gedamu at gmail.com (Abraham Gedamu) Date: Thu, 27 Jul 2006 10:51:07 -0400 Subject: [MINC-users] Registration using only translation and scale in minc In-Reply-To: References: Message-ID: Hey Andrew, Running minctracc: minctracc -lsq9 -w_rotations 0 0 0 source.mnc.gz target.mnc.gzno_rotation.xfm COG of v1: 60.239826 68.026634 58.329350 COG of v2: 61.871235 58.958786 59.136723 [trans] = 1.631409 -9.067848 0.807373 Initial objective function val = 0.07533003 Final objective function value = 0.05163525 more no_rotation.xfm Transform_Type = Linear; Linear_Transform = 0.974233269691467 -0.106358632445335 0.116122744977474 4.05126190185547 0.0418491959571838 0.463993549346924 0.0738773718476295 22.1566772460938 -0.139944493770599 -0.152131617069244 1.03475058078766 7.68611145019531; xfm2param no_rotation.xfm after parameter extraction -center 0.00000 0.00000 0.00000 -translation 4.05126 22.15668 7.68611 -rotation -8.36385 7.62126 5.13547 -scale 0.98688 0.47170 1.05520 -shear 0.00000 0.00000 0.00000 Abe On 7/26/06, Abraham Gedamu wrote: > > I tried minctracc -lsq9 w_rotations 0 0 0 source.mnc target.mnc out.xfmseveral times with different source files and target files, however when I > run xfm2param on the out.xfm file, my rotations are not 0. Does anyone no > why this might occur? > > Thanks, > > Abe > > > > On 7/25/06, Abraham Gedamu wrote: > > > > Hello, > > > > I would like to register one image to another using only translation and > > scale. I have tried using minctracc but there are no options that allow the > > use of only translation and scale. I could get all the parameters and set > > the ones I don't want to zero, but then the values for scale and translation > > will not be the correct ones since the paramters seem to depend on one > > another. Any advice in this are would be greatly appreciated. > > > > Thanks > > > > Abe > > > > From a.janke at gmail.com Thu Jul 27 11:04:04 2006 From: a.janke at gmail.com (Andrew Janke) Date: Thu, 27 Jul 2006 11:04:04 -0400 Subject: [MINC-users] Registration using only translation and scale in minc In-Reply-To: References: Message-ID: > minctracc -lsq9 -w_rotations 0 0 0 source.mnc.gz target.mnc.gzno_rotation.xfm > COG of v1: 60.239826 68.026634 58.329350 > COG of v2: 61.871235 58.958786 59.136723 > [trans] = 1.631409 -9.067848 0.807373 > Initial objective function val = 0.07533003 > Final objective function value = 0.05163525 > > more no_rotation.xfm > Transform_Type = Linear; > Linear_Transform = > 0.974233269691467 -0.106358632445335 0.116122744977474 4.05126190185547 > 0.0418491959571838 0.463993549346924 0.0738773718476295 22.1566772460938 > -0.139944493770599 -0.152131617069244 1.03475058078766 7.68611145019531; > > xfm2param no_rotation.xfm > after parameter extraction > -center 0.00000 0.00000 0.00000 > -translation 4.05126 22.15668 7.68611 > -rotation -8.36385 7.62126 5.13547 > -scale 0.98688 0.47170 1.05520 > -shear 0.00000 0.00000 0.00000 Ah! now it begins to make sense.. minctracc works by a two step process, if no initial transformation is given (I had imagined you were starting with identity or something). If no initial transformation is given (be that via -transformation or -identity) then first a PAT transformation is estimated. From there a simplex optimisation of this initial guess/stab in the dark is performed. Thus you also will have to restrict the initial PAT transformation (or start from -identity). This should work: minctracc -est_center -est_scale -est_translations \ -lsq9 -w_rotations 0 0 0 \ source.mnc target.mnc out.xfm a From abraham.gedamu at gmail.com Thu Jul 27 11:22:27 2006 From: abraham.gedamu at gmail.com (Abraham Gedamu) Date: Thu, 27 Jul 2006 11:22:27 -0400 Subject: [MINC-users] Registration using only translation and scale in minc In-Reply-To: References: Message-ID: That did the trick. Thanks for the help. Abe On 7/27/06, Abraham Gedamu wrote: > > Hey Andrew, > > Running minctracc: > > minctracc -lsq9 -w_rotations 0 0 0 source.mnc.gz target.mnc.gzno_rotation.xfm > COG of v1: 60.239826 68.026634 58.329350 > COG of v2: 61.871235 58.958786 59.136723 > [trans] = 1.631409 -9.067848 0.807373 > Initial objective function val = 0.07533003 > Final objective function value = 0.05163525 > > more no_rotation.xfm > Transform_Type = Linear; > Linear_Transform = > 0.974233269691467 -0.106358632445335 0.116122744977474 4.05126190185547 > 0.0418491959571838 0.463993549346924 0.0738773718476295 22.1566772460938 > -0.139944493770599 -0.152131617069244 1.03475058078766 7.68611145019531; > > xfm2param no_rotation.xfm > after parameter extraction > -center 0.00000 0.00000 0.00000 > -translation 4.05126 22.15668 7.68611 > -rotation -8.36385 7.62126 5.13547 > -scale 0.98688 0.47170 1.05520 > -shear 0.00000 0.00000 0.00000 > > Abe > > > On 7/26/06, Abraham Gedamu wrote: > > > > I tried minctracc -lsq9 w_rotations 0 0 0 source.mnc target.mnc out.xfmseveral times with different source files and target files, however when I > > run xfm2param on the out.xfm file, my rotations are not 0. Does anyone > > no why this might occur? > > > > Thanks, > > > > Abe > > > > > > > > On 7/25/06, Abraham Gedamu < abraham.gedamu at gmail.com> wrote: > > > > > > Hello, > > > > > > I would like to register one image to another using only translation > > > and scale. I have tried using minctracc but there are no options that allow > > > the use of only translation and scale. I could get all the parameters and > > > set the ones I don't want to zero, but then the values for scale and > > > translation will not be the correct ones since the paramters seem to depend > > > on one another. Any advice in this are would be greatly appreciated. > > > > > > Thanks > > > > > > Abe > > > > > > > > From a.janke at gmail.com Fri Jul 28 02:49:24 2006 From: a.janke at gmail.com (Andrew Janke) Date: Fri, 28 Jul 2006 02:49:24 -0400 Subject: [MINC-users] Question with mincedit In-Reply-To: <005c01c6b189$803af060$c32914c6@irus.robarts.ca> References: <005c01c6b189$803af060$c32914c6@irus.robarts.ca> Message-ID: Hi Jessie, If you could send the exact error you are having I will be able to provide a more sensible answer, however I have seen problems with large dicom headers being stuffed into the MINC header. You can use the attached script to ameliorate this. Just be aware that the script works on the file inplace. a On 7/27/06, (Jessie) Ting Guo wrote: > Hi, > > I'm trying to edit the head file of a minc image using mincedit. However I got the memory shortage problem no matter which editor I was using. The RAM on my desktop is about 2G. Could anyone tell me how to fix this? > > Thank you very much! > > Jessie > _______________________________________________ > MINC-users at bic.mni.mcgill.ca > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users > -- Andrew Janke (a.janke at gmail.com || http://a.janke.googlepages.com/) Canada->Montreal Cell: +1 (514) 924 2012 -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: minc_de-dicomerise Url: http://www.bic.mni.mcgill.ca/pipermail/minc-users/attachments/20060728/4a44402d/minc_de-dicomerise.ksh From pgravel at bic.mni.mcgill.ca Fri Jul 28 12:37:58 2006 From: pgravel at bic.mni.mcgill.ca (Paul GRAVEL) Date: Fri, 28 Jul 2006 12:37:58 -0400 Subject: [MINC-users] Minctools on MAC OS 10.3 Message-ID: Dear All, I need to install the minctools on a MAC running OS 10.3.9. I am downloading the file from: http://packages.bic.mni.mcgill.ca/osx-10.3/ Once the installation of the packages is done, should everything be running smoothly, or do I need to install other packages, e.g. GLUT, before being able to run programs such as Display? Thanking you in advance. Best Regards, Paul From a.janke at gmail.com Fri Jul 28 13:50:23 2006 From: a.janke at gmail.com (Andrew Janke) Date: Fri, 28 Jul 2006 13:50:23 -0400 Subject: [MINC-users] Minctools on MAC OS 10.3 In-Reply-To: References: Message-ID: > I need to install the minctools on a MAC running OS 10.3.9. I am > downloading the file from: > http://packages.bic.mni.mcgill.ca/osx-10.3/ > > Once the installation of the packages is done, should everything be > running smoothly, or do I need to install other packages, e.g. GLUT, > before being able to run programs such as Display? There are still a few things that have to be done by hand. :( THis includes: 1. Install X11 (if not already -- usually it is). 2. install netpbm via fink 3. install mni_perllib in the usual fashion 4. install GetOpt::Tabular from CPAN. Besides that, update your PATH and things should be good. a