[Freesurfer] Reanalyzing some OASIS cases
Hi, I would like to rerun recon-all (from the latest freesurfer) on a few cases from OASIS brains study, to get control values for my patient group. Since my computational options are rather limitted, I would like to avoid running redundant steps, which would anyway give me the same results, or to repeat possible surface editing. Anybody could please suggest me where to start the analysis or in general how to tackle this problem? Thanks in advance, Martin ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] Talairach_avi and eTIV, version 3.0.5 vs 4.0.5
Hi, for those interested: we have correlated FS's ICV estimates and estimates from a procedure developed by Anders Dale (a deformable template procedure, similar to the Shrink Wrapping procedure previously described by Dale and colleagues (Dale et al, 1999; Dale and Sereno, 1993)). Mean ICV from FS: 1567690 (SD = 161247) Mean ICV from Dale: 1584008 (SD = 136979) The difference was not significant (p = .12, t = -1.596) The correlation was .84. A scatterplot is attached (FS on the y-axis, Dale's method on the x). Not too bad, but not perfect either. The sample was a life-span sample (20-90 years, n = 74). Hope this info was usefull. Best, Anders MF > > Hi Paul, > Just as an aside, once you have a talairach_with_skull.lta file (whether > from v3 or v4), you can calculate eTIV as scale_factor/determinant(lta), > where the scale_factor employed is 2150. > > Incidentally, using the -eTIV option of mri_label_volume, you can also > specify other transforms (e.g., talairach.xfm, talairach.lta) to be used > used instead of the default talairach_with_skull.lta in computing "ICV". > I've looked at the effect of using different transforms in a handful of > brains, and "ICV" can vary a bit depending on transform you choose to > use. > > I'm curious: Has anyone out there compared FS's ICV estimate to ICV > estimated by another program? > > Best, > Mike H. > > > On Fri, 2008-06-20 at 14:35 -0700, Paul Greenberg wrote: >> Hello Freesurfers, >> >> I am interested in comparing eTIV calculated with version 3.0.5 to >> that calculated by version 4.0.5 (64 bit Cent OS). >> >> In order to compare eTIV for a single brain estimated in version 3.0.5 >> to version 4.0.5, I copied a subject's previous freesurfer output >> (v3.0.5) into the freesurfer version 4.0.5 folder, deleted all the >> files within the "transforms" folder and carried out the following >> procedures as outlined in the recon-all dev table: >> >> 1) Run the talairach procedure. Command = talairach_avi --i >> mysubject --xfm talairach.auto.xfm >> 2) cp talairach.auto.xfm talairach.xfm >> 3) Run automatic failure detection. Command= talairach_afd -T 0.005 >> -xfm talairach.xfm >> 4) Create talairach_with_skull.lta. Command = >> mri_em_register -skull nu.mgz >> $freesurferhome/average/RB_all_withskull_2007-08-08.gca >> talairach_with_skull.lta >> 5) Check the new eTIV. Command = >> mri_label_volume -eTIV talairach_with_skull.lta aseg.mgz 17 53 >> As the aseg.mgz file exists from previous processing (versio >> 3.0.5), and the talairach_with_skull.lta was re-created with the new >> processing (version 4.0.5), the output of this command should list the >> eTIV based on the new talairach transform? >> >> Are these procedures correct for generating new values of eTIV if that >> is all i want? >> >> Thanks! >> >> -- >> >> Paul Greenberg, Ph.D. >> Radiology Department, University of California, San Diego >> VA San Diego Healthcare System (116A-13) >> 3350 La Jolla Village Drive >> San Diego, CA 92161 >> Phone: (858) 552-8585 ext. 2903 >> Fax: (858) 642-3836 >> [EMAIL PROTECTED] >> ___ >> Freesurfer mailing list >> Freesurfer@nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > ___ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > ICV_scatters.pdf Description: Adobe PDF document ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
[Freesurfer] inorm-sess error
Hi, I'm trying to get around the problems from my previous post: http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg07669.html I am now trying to use mri_convert to convert from nifti to bshort using the following command: mri_convert -i rfkoch-0021-00144-000144-00.nii -it nii -o fspm2_000.bshort -ot bshort The bshort files seem to be created but then when I run inorm-sess I get the following error. The odd thing is that when I used mri_concat on exactly the same .nii files (as mentioned in my previous post) I was able to run inorm-sess with no errors... What is the difference between these two commands? The inorm error: Warning: Matrix is singular to working precision. > In fast_inorm at 196 ??? Error using ==> betainc at 35 X must be in the interval [0,1]. The exact output from the command is pasted below. Here is the output of bugr: - FREESURFER_HOME: /Applications/freesurfer Build stamp: freesurfer-Darwin-tiger-i686-stable-pub-v3.0.5 Kernel info: Darwin 8.11.1 i386 - If anyone has any suggestions I'd be grateful to hear them! Thanks! inorm-sess -sf sessid -df sessdir -rlf runlist_inormcheck -funcstem fspm2 /Volumes/donnees/Imagery/imagery_data/RF005 Thu Jun 19 17:43:17 CEST 2008 Filesystem 512-blocks Used Avail Capacity Mounted on /dev/disk0s3 808067040 465870224 34219681658%/Volumes/donnees RunList 021 -- /Volumes/donnees/Imagery/imagery_data/RF005/bold inorm -i 021/fspm2 -umask 22 -TR 2. -ipr 3 -bpr 2.998958184 -seqname PerImg -- Log file is 021/inorm.log inorm matlab file is /tmp/inorm_21280.m --- Matlab file -- %%% - %% % matlab file to run inorm % This is a temporary file and may be deleted % /tmp/inorm_21280.m global QuitOnError; QuitOnError = 1; r = fast_inorm( '-i', '021/fspm2', '-umask', '22', '-TR', '2.', '-ipr', '3', '-bpr', '2.998958184', ' ', 'PerImg'); quit; --- -- --- matlab output Warning: Unable to open display iconic, MATLAB is starting without a display. You will not be able to display graphics on the screen. < M A T L A B > Copyright 1984-2007 The MathWorks, Inc. Version 7.4.0.287 (R2007a) January 29, 2007 Warning: Duplicate directory name: /Users/lreddy/matlab. To get started, type one of these: helpwin, helpdesk, or demo. For product information, visit www.mathworks.com. >> >> >> >> >> >> >> $Id Parsing Arguments Checking Parameters invol021/fspm2 firstslice 0 nslices 32 relthresh0.75 reportfile 021/fspm2.report meanvalfile 021/fspm2.meanval twfstem 021/fspm2.twf TR 2.00 First Pass: Computing global mean 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Second Pass: Segmenting 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Warning: Matrix is singular to working precision. > In fast_inorm at 125 Warning: Matrix is singular to working precision. > In fast_inorm at 146 Warning: Matrix is singular to working precision. > In fast_inorm at 175 Warning: Matrix is singular to working precision. > In fast_inorm at 196 ??? Error using ==> betainc at 35 X must be in the interval [0,1]. Error in ==> tTest at 37 p = betainc(z,dof/2,0.5); Error in ==> fast_inorm at 223 tSigCor = tTest(s.ntp-2,tCor); >> 2008-06-19 17:43:31.661 MATLAB[21307] Process manager already initialized >> -- can't fully enable headless m ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] Talairach_avi and eTIV, version 3.0.5 vs 4.0.5
Hi yes, the FS ICVs mostly seem to work :-) I still think one should watch out for outliers. We had a sample of 28 kids aged 8-13 yrs, and the ICVs were quite a bit off and useless for two of those, so I would check all carefully especially if you have a group that may have smaller than normal adult ICV. Best, K > Hi, for those interested: we have correlated FS's ICV estimates and > estimates from a procedure developed by Anders Dale (a deformable template > procedure, similar to the Shrink Wrapping procedure previously described > by Dale and colleagues (Dale et al, 1999; Dale and Sereno, 1993)). > > Mean ICV from FS: 1567690 (SD = 161247) > Mean ICV from Dale: 1584008 (SD = 136979) > > The difference was not significant (p = .12, t = -1.596) > > The correlation was .84. A scatterplot is attached (FS on the y-axis, > Dale's method on the x). > > Not too bad, but not perfect either. > > The sample was a life-span sample (20-90 years, n = 74). > > Hope this info was usefull. > > Best, > Anders MF > > > > >> >> Hi Paul, >> Just as an aside, once you have a talairach_with_skull.lta file (whether >> from v3 or v4), you can calculate eTIV as scale_factor/determinant(lta), >> where the scale_factor employed is 2150. >> >> Incidentally, using the -eTIV option of mri_label_volume, you can also >> specify other transforms (e.g., talairach.xfm, talairach.lta) to be used >> used instead of the default talairach_with_skull.lta in computing "ICV". >> I've looked at the effect of using different transforms in a handful of >> brains, and "ICV" can vary a bit depending on transform you choose to >> use. >> >> I'm curious: Has anyone out there compared FS's ICV estimate to ICV >> estimated by another program? >> >> Best, >> Mike H. >> >> >> On Fri, 2008-06-20 at 14:35 -0700, Paul Greenberg wrote: >>> Hello Freesurfers, >>> >>> I am interested in comparing eTIV calculated with version 3.0.5 to >>> that calculated by version 4.0.5 (64 bit Cent OS). >>> >>> In order to compare eTIV for a single brain estimated in version 3.0.5 >>> to version 4.0.5, I copied a subject's previous freesurfer output >>> (v3.0.5) into the freesurfer version 4.0.5 folder, deleted all the >>> files within the "transforms" folder and carried out the following >>> procedures as outlined in the recon-all dev table: >>> >>> 1) Run the talairach procedure. Command = talairach_avi --i >>> mysubject --xfm talairach.auto.xfm >>> 2) cp talairach.auto.xfm talairach.xfm >>> 3) Run automatic failure detection. Command= talairach_afd -T 0.005 >>> -xfm talairach.xfm >>> 4) Create talairach_with_skull.lta. Command = >>> mri_em_register -skull nu.mgz >>> $freesurferhome/average/RB_all_withskull_2007-08-08.gca >>> talairach_with_skull.lta >>> 5) Check the new eTIV. Command = >>> mri_label_volume -eTIV talairach_with_skull.lta aseg.mgz 17 53 >>> As the aseg.mgz file exists from previous processing (versio >>> 3.0.5), and the talairach_with_skull.lta was re-created with the new >>> processing (version 4.0.5), the output of this command should list the >>> eTIV based on the new talairach transform? >>> >>> Are these procedures correct for generating new values of eTIV if that >>> is all i want? >>> >>> Thanks! >>> >>> -- >>> >>> Paul Greenberg, Ph.D. >>> Radiology Department, University of California, San Diego >>> VA San Diego Healthcare System (116A-13) >>> 3350 La Jolla Village Drive >>> San Diego, CA 92161 >>> Phone: (858) 552-8585 ext. 2903 >>> Fax: (858) 642-3836 >>> [EMAIL PROTECTED] >>> ___ >>> Freesurfer mailing list >>> Freesurfer@nmr.mgh.harvard.edu >>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> >> ___ >> Freesurfer mailing list >> Freesurfer@nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> > ___ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
[Freesurfer] white matter volume - version 3.0.5 vs 4.0.5
Dear Freesurfers, Is it acceptable to use the program "mris_wm_volume" from version 4.0.5 (Cent OS, x86_64) to calculate total white matter volume from data processed with version 3.0.5? Would I need to subtract any subcortical volumes such as ventricles, basal ganglia, etc, from the output of mris_wm_volume (version 4.0.5) in order to get the most accurate measurement of total white matter volume? Thanks! Paul -- Paul Greenberg, Ph.D. Radiology Department, University of California, San Diego VA San Diego Healthcare System (116A-13) 3350 La Jolla Village Drive San Diego, CA 92161 Phone: (858) 552-8585 ext. 2903 Fax: (858) 642-3836 [EMAIL PROTECTED] ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
[Freesurfer] GRF correction
Hi, I understand that there are tools under development implementing Gaussian Random Field correction for multiple comparisons. If there is a usable version available (and a brief description of how to use it), I'd like to try it out. Thanks in advance, Cate Hartley ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] GRF correction
You mena for FDR estimation? 2008/6/23 Catherine Hartley <[EMAIL PROTECTED]>: > Hi, > > I understand that there are tools under development implementing Gaussian > Random Field correction for multiple comparisons. If there is a usable > version available (and a brief description of how to use it), I'd like to > try it out. > > Thanks in advance, > Cate Hartley > ___ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] GRF correction
You mean for FDR estimation? 2008/6/23 Pedro Paulo de Magalhães Oliveira Junior <[EMAIL PROTECTED]>: > You mena for FDR estimation? > > > > 2008/6/23 Catherine Hartley <[EMAIL PROTECTED]>: > > Hi, >> >> I understand that there are tools under development implementing Gaussian >> Random Field correction for multiple comparisons. If there is a usable >> version available (and a brief description of how to use it), I'd like to >> try it out. >> >> Thanks in advance, >> Cate Hartley >> ___ >> Freesurfer mailing list >> Freesurfer@nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> > > ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] Talairach_avi and eTIV, version 3.0.5 vs 4.0.5
Hello All, We are interested in the ranges of eTIV values found by other Freesurfer users. As noted below, one research group reports a mean ICV of about 1560 cc for adults ranging from 20-90 years old. Our sample (n=40) including males and females (51 to 83 years old) has a mean eTIV of 1520 cc (standard deviation = 200 cc), and a median eTIV of 1511 cc. Do these values seem reasonable? Thanks, Paul [EMAIL PROTECTED] wrote: Hi, for those interested: we have correlated FS's ICV estimates and estimates from a procedure developed by Anders Dale (a deformable template procedure, similar to the “Shrink Wrapping” procedure previously described by Dale and colleagues (Dale et al, 1999; Dale and Sereno, 1993)). Mean ICV from FS: 1567690 (SD = 161247) Mean ICV from Dale: 1584008 (SD = 136979) The difference was not significant (p = .12, t = -1.596) The correlation was .84. A scatterplot is attached (FS on the y-axis, Dale's method on the x). Not too bad, but not perfect either. The sample was a life-span sample (20-90 years, n = 74). Hope this info was usefull. Best, Anders MF Hi Paul, Just as an aside, once you have a talairach_with_skull.lta file (whether from v3 or v4), you can calculate eTIV as scale_factor/determinant(lta), where the scale_factor employed is 2150. Incidentally, using the -eTIV option of mri_label_volume, you can also specify other transforms (e.g., talairach.xfm, talairach.lta) to be used used instead of the default talairach_with_skull.lta in computing "ICV". I've looked at the effect of using different transforms in a handful of brains, and "ICV" can vary a bit depending on transform you choose to use. I'm curious: Has anyone out there compared FS's ICV estimate to ICV estimated by another program? Best, Mike H. On Fri, 2008-06-20 at 14:35 -0700, Paul Greenberg wrote: Hello Freesurfers, I am interested in comparing eTIV calculated with version 3.0.5 to that calculated by version 4.0.5 (64 bit Cent OS). In order to compare eTIV for a single brain estimated in version 3.0.5 to version 4.0.5, I copied a subject's previous freesurfer output (v3.0.5) into the freesurfer version 4.0.5 folder, deleted all the files within the "transforms" folder and carried out the following procedures as outlined in the recon-all dev table: 1) Run the talairach procedure. Command = talairach_avi --i mysubject --xfm talairach.auto.xfm 2) cp talairach.auto.xfm talairach.xfm 3) Run automatic failure detection. Command= talairach_afd -T 0.005 -xfm talairach.xfm 4) Create talairach_with_skull.lta. Command = mri_em_register -skull nu.mgz $freesurferhome/average/RB_all_withskull_2007-08-08.gca talairach_with_skull.lta 5) Check the new eTIV. Command = mri_label_volume -eTIV talairach_with_skull.lta aseg.mgz 17 53 As the aseg.mgz file exists from previous processing (versio 3.0.5), and the talairach_with_skull.lta was re-created with the new processing (version 4.0.5), the output of this command should list the eTIV based on the new talairach transform? Are these procedures correct for generating new values of eTIV if that is all i want? Thanks! -- Paul Greenberg, Ph.D. Radiology Department, University of California, San Diego VA San Diego Healthcare System (116A-13) 3350 La Jolla Village Drive San Diego, CA 92161 Phone: (858) 552-8585 ext. 2903 Fax: (858) 642-3836 [EMAIL PROTECTED] ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer -- Paul Greenberg, Ph.D. Radiology Department, University of California, San Diego VA San Diego Healthcare System (116A-13) 3350 La Jolla Village Drive San Diego, CA 92161 Phone: (858) 552-8585 ext. 2903 Fax: (858) 642-3836 [EMAIL PROTECTED] ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] GRF correction
We do not have a GRF algorithm (at least one that I trust yet). Alternatively, you can perform monte carlo or permutation simulation with mri_glmfit, and this gives the same (or similar) information. > You mean for FDR estimation? > > 2008/6/23 Pedro Paulo de Magalhães Oliveira Junior <[EMAIL PROTECTED]>: > >> You mena for FDR estimation? >> >> >> >> 2008/6/23 Catherine Hartley <[EMAIL PROTECTED]>: >> >> Hi, >>> >>> I understand that there are tools under development implementing >>> Gaussian >>> Random Field correction for multiple comparisons. If there is a usable >>> version available (and a brief description of how to use it), I'd like >>> to >>> try it out. >>> >>> Thanks in advance, >>> Cate Hartley >>> ___ >>> Freesurfer mailing list >>> Freesurfer@nmr.mgh.harvard.edu >>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>> >> >> > ___ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] conjoining labels
try mri_mergelabels (or mri_merge_lables, cant remember which). > Hi, > > I can't seem to figure out how to save two or more labels as a > conjoined single label. > > Thanks in advance, > -Cate > ___ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer