Freesurfer November 2014

freesurfer@nmr.mgh.harvard.edu

99 participants
144 discussions

Speeding Freesurfer
by Francesco Cardinale 27 Nov '14

27 Nov '14

Dear Freesurfer experts, I've been using Freesurfer on Mac computers for some years with great satisfaction. I just bought a "monster computer", a Mac Pro with 12-core Xeon E5, 64 Gb RAM, GPU AMD FirePro D700. This is, out of doubt, a marvellous machine to compute many subjects simultaneously, but quite often we need to run one single subject as fast as possible. What can I do? OpenMP optimization is available only for CentOS, isn't it? What about CentOS in a VM, will it be able to exploit multithreading, or is it necessary to configure a multiboot machine with BootCamp? Any suggestions to do that? Moreover, what about CUDA and GPU acceleration? Is it available for FirePro graphic cards? Any other suggestions? Anything really new in the field with the next version of Freesurfer? Thanks a lot Francesco Cardinale, MD Neurosurgeon "Claudio Munari" Centre for Epilepsy and Parkinson Surgery - Ospedale Niguarda "Ca' Granda" Piazza dell'Ospedale Maggiore, 3 - 20162 - Milano - Italia phone 0039 02 64442917 fax 0039 02 64442868 e-mail francesco.cardinale(a)ospedaleniguarda.it

1 0

mri_glmfit error
by Bronwyn Overs 26 Nov '14

26 Nov '14

Dear Freesurfer Mailing list, I have been attempting to run the following command in freesurfer v5.3.0: mri_glmfit --y site-gender-group-prothaplotype.thick.lh.10.mgh --fsgd site-gender-group-prothaplotype.fsgd dods --C me_site-BrisVsMelb.mtx --C me_site-BrisVsPer.mtx --C me_site-MelbVsPer.mtx --C --surf fsaverage lh --cortex --glmdir site-gender-group-prothaplotype.thick.lh.glmdir Unfortunely, I keep getting the following error: "ERROR: Option lh unknown" Do you know why this would be? Kind regards, Bronwyn Overs Research Assistant Neuroscience Research Australia Neuroscience Research Australia Margarete Ainsworth Building Barker Street Randwick Sydney NSW 2031 Australia *M* 0411 308 769 *T* +61 2 9399 1883 *F* +61 2 9399 1265 neura.edu.au <http://neura.edu.au> Follow @neuraustralia on twitter <https://twitter.com/neuraustralia>Follow NeuRA on facebook <https://www.facebook.com/NeuroscienceResearchAustralia>Subscribe to the NeuRA Magazine <http://www.neura.edu.au/help-research/subscribe>

1 0

ERROR:dimension mismatch between X and contrast con.mtx
by Xinfa Shi 26 Nov '14

26 Nov '14

Hi everyone, I get this error when running group analysis of cortical thickness by command-line: xin-fa@xinfa--PC:/media/E/FREESURFER/VisualData$ mri_glmfit --y lh.group_differ.thickness.10.mgh --fsgd group_differ.fsgd dods --C con.mtx --surf fsaverage lh --cortex --glmdir lh.group_differ.glmdir gdfReadHeader: reading group_differ.fsgd INFO: gd2mtx_method is dods Reading source surface /media/E/FREESURFER/VisualData/fsaverage/surf/lh.white Number of vertices 163842 Number of faces 327680 Total area 65416.648438 AvgVtxArea 0.399267 AvgVtxDist 0.721953 StdVtxDist 0.195470 $Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /media/E/FREESURFER/VisualData cmdline mri_glmfit --y lh.group_differ.thickness.10.mgh --fsgd group_differ.fsgd dods --C con.mtx --surf fsaverage lh --cortex --glmdir lh.group_differ.glmdir sysname Linux hostname xinfa--PC machine x86_64 user xin-fa FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /media/E/FREESURFER/VisualData/lh.group_differ.thickness.10.mgh logyflag 0 usedti 0 FSGD group_differ.fsgd labelmask /media/E/FREESURFER/VisualData/fsaverage/label/lh.cortex.label maskinv 0 glmdir lh.group_differ.glmdir IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory lh.group_differ.glmdir Loading y from /media/E/FREESURFER/VisualData/lh.group_differ.thickness.10.mgh INFO: gd2mtx_method is dods Saving design matrix to lh.group_differ.glmdir/Xg.dat Normalized matrix condition is 1 Matrix condition is 1.04762 Found 149955 points in label. Pruning voxels by thr: 0.000000 Found 149953 voxels in mask Saving mask to lh.group_differ.glmdir/mask.mgh Reshaping mriglm->mask... search space = 74612.446501 ERROR: dimension mismatch between X and contrast con.mtx X has 2 cols, C has 1 cols The data is altogether 43 subjects,including 22 subjects in expert group.I just want to explore the difference between both groups. Attached is the FSGD file ,which is two groups ,zero covariates . The contrast matrix is 1 -1 . Subsequently I try it only with a few subjects ,and it succeeds. Can you help me ? Best wishes. Xin-Fa .

2 1

prefrontal cortex label
by Anna Jonsson 26 Nov '14

26 Nov '14

hi, I have tried to make prefrotnal cortex label by doing first mri_annotation2label --subject "$subject" --hemi rh --outdir "$subject"/label mri_mergelabels -i lh.caudalanteriorcingulate.label -i lh.caudalmiddlefrontal.label -i lh.lateralorbitofrontal.label -i lh.medialorbitofrontal.label -i lh.parsopercularis.label -i lh.parsorbitalis.label -i lh.parstriangularis.label -i lh.rostralanteriorcingulate.label -i lh.rostralmiddlefrontal.label -i lh.superiorfrontal.label -i lh.lateralorbitofrontal.label -o lh.prefrontalcortex.label -d /directory/subject/label which does create a label called x.prefrontalcortex.label in each persons directory. then however when i try to do mris_anatomical_stats -l $SUBJECTS_DIR/"$subject"/label/lh.prefrontalcortex.label "$subject" lh pial | grep "thickness" >> left_prefrontal i get error and say it cant find label. any pointers on what im doing wrong? thank you

3 3

Different significant regions with different cluster wise thresholds
by Bronwyn Overs 26 Nov '14

26 Nov '14

Dear Freesurfer Mailing list, While running a glm anlaysis with monte carlo correction (using cluster wise thresholds of .05 & .01), I encountered the following issue: When examining the clusters formed for each of the different cluster wise thresholds, the .05 threshold resulted in a superior temporal cluster, while the .01 threshold resulted in a two lateral occipital clusters that did not overlap with the cluster from the .05 threshold. Do you know why there would be no overlap between significant regions at different cluster wise thresholds? I was aware that a cluster which is significant at .05 may split or be smaller at .01, but I do not know why entirely different regions would be significant. -- Kind regards, Bronwyn Overs Research Assistant Neuroscience Research Australia Neuroscience Research Australia Margarete Ainsworth Building Barker Street Randwick Sydney NSW 2031 Australia *M* 0411 308 769 *T* +61 2 9399 1883 *F* +61 2 9399 1265 neura.edu.au <http://neura.edu.au> Follow @neuraustralia on twitter <https://twitter.com/neuraustralia>Follow NeuRA on facebook <https://www.facebook.com/NeuroscienceResearchAustralia>Subscribe to the NeuRA Magazine <http://www.neura.edu.au/help-research/subscribe>

3 6

mri_glmfit-sim
by Jürgen Hänggi 26 Nov '14

26 Nov '14

Dear FS experts We would like to force FS's mri_glmfit-sim function of FS 5.3.0 to report all results and not only the significant results. This was the behavior of this function until FS version 4.5.0. Please can you guide as to the relevant lines in the mri_glmfit-sim function and indicate what have to be changed. Thanks in advance Best regards Jürgen Hänggi ---------------------------------------------------------------------------- Jürgen Hänggi, Ph.D. Division Neuropsychology Institute of Psychology University of Zurich Binzmuehlestrasse 14, PO Box 25 8050 Zurich, Switzerland 0041 44 635 73 97 (phone office) 0041 76 445 86 84 (phone mobile) 0041 44 635 74 09 (fax office) BIN 4.D.04 (office room number) j.haenggi[at]psychologie.uzh.ch (email) http://www.psychologie.uzh.ch/neuropsy/ (website) http://www.juergenhaenggi.ch (private website) This e-mail (and any attachment/s) contains confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden. ----------------------------------------------------------------------------

2 1

parcellation of group stats maps?
by Scott Hayes 26 Nov '14

26 Nov '14

Hi, Is there a way to generate an intersection between a cortical thickness stats map (e.g., sig.mgh) in fsaverage space and native space neuroanatomy using FS parcellations (e.g., lh.apac.anot.)? We would like to plot cortical thickness of the significant clusters for each subject, but separate the significant clusters by the parcellations, e.g., if the significant cluster spans parahippocampal and fusiform cortex, I would like those plotted separately (preferably w/o drawing ROIs--and again, not using the entire PHC or fusiform parcellation--only the portion that is significant in the group analysis). It seems like this might be possible to accomplish w/ a single command, but we have not been able to pinpoint the exact command w/ the necessary flags. Any guidance would be appreciated. Thanks, Scott

2 1

Re: [Freesurfer] Fwd: Eccentric lesions (fwd)
by Lilla Zollei 26 Nov '14

26 Nov '14

Hi Octavian, If you have an OK aseg volume, you could try to fit a surface over it using mris_make_surfaces with the -cover_seg flag. Lilla On Tue, 25 Nov 2014, Bruce Fischl wrote: > > > ---------- Forwarded message ---------- > Date: Tue, 25 Nov 2014 08:05:55 -0600 > From: Octavian Lie <octavian.lie(a)gmail.com> > Reply-To: Freesurfer support list <freesurfer(a)nmr.mgh.harvard.edu> > To: Freesurfer support list <freesurfer(a)nmr.mgh.harvard.edu> > Subject: Re: [Freesurfer] Fwd: Eccentric lesions > > It looks like there is only gray matter (int 42), similar to cortical ribbon > gray, no white matter in that gyrus in aseg, both in the original recon, and > after painting gyral centers in brainmask. > Octavian > > On Tue, Nov 25, 2014 at 7:51 AM, Octavian Lie <octavian.lie(a)gmail.com> > wrote: > Dear Bruce, > > I am not sure how an ok aseg.mgz would look like for a lesion like > that, but it looks like it includes one larger gyrus excluded by the > pia, see attached, I would be happy to retrieve that one as it is the > bulkiest of all the ones left out. > Thank you, > > Octavian > > On Tue, Nov 25, 2014 at 7:38 AM, Bruce Fischl > <fischl(a)nmr.mgh.harvard.edu> wrote: > Hi Octavian > > I wouldn't use control points since it really isn't > healthy wm. The only > thing you can do is edit the wm.mgz and possibly also the > brainmask, but > it's so abnormal looking you may not be able to get > accurate surfaces. > How does the aseg look? If it is ok there may be one other > option. > > cheers > Bruce > > > On Tue, 25 Nov 2014, Octavian Lie wrote: > > > Dear All, > > any suggestion about this type of problem? I tried > control points through > > the middle/axis of the atrophic gyri (intensity 30-90 in > T1 or brainmask), > > wm.mgz edits through the same (adding from brainmask by > cloning those > > voxels; painting 110; or 255 brush), and modifying > brainmask directly by > > painting 110 internsity voxels through the same (gyri > interiors, where wm is > > supposed to be). Did not work, the atrophic gyri > continue to remain outside > > the pia. Again, this is an example of an eccentric > lesion with gyral > > atrophy, without interruption in the gray matter ribbon > but very faint wm > > signal through the base of the affected gyri. > > Thanks, > > Octavian > > > > > > > > ---------- Forwarded message ---------- > > From: Octavian Lie <octavian.lie(a)gmail.com> > > Date: Tue, Nov 18, 2014 at 10:50 AM > > Subject: Eccentric lesions > > To: "freesurfer(a)nmr.mgh.harvard.edu" > <freesurfer(a)nmr.mgh.harvard.edu> > > > > > > Dear All, > > > > I am interested at getting as good pial/cortical > reconstructions as possible > > on a series of epilepsy pt scans, some showing eccentric > (cortical) lesions > > (focal encephalomalacias, infarcts, or after focal > resections). One > > particularly difficult lesion type to get an accurate > cortical surface is > > where there are prongs of grey matter bulging out (like > a bag of worms), > > (mostly) without visible wm, see attached picture. These > prongs are > > maintained in the brainmask.mgz and > brain.finalsurfs.mgz, but not included > > in the pia with the default recon. Lesion voxel > intensities vary from 30-90. > > I tried cp, wm edits (using brainmask or T1 as > referece), both, did not > > matter, it did not work. > > > > Since I am not intested in wm/subcortical segmentation > but just in a good > > pia, I was wondering if I 'create' thin wm tracts > through the center of > > these gm prongs to help pial segmentation. If this is a > valid option, should > > I create those in brainmask, brain.finalsurfs or wm.mgz, > and should I use > > 110 or 255 for the brush? > > Any other suggestions are appreciated. Nevertheless, I > know there is a lot > > of tweaking of each lesion, and I try for more of a > global approach to these > > kinds of lesions. > > > > Thank you, > > > > Octavian. > > > > > > > _______________________________________________ > Freesurfer mailing list > Freesurfer(a)nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > The information in this e-mail is intended only for the person > to whom it is > addressed. If you believe this e-mail was sent to you in error > and the e-mail > contains patient information, please contact the Partners > Compliance HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent > to you in error > but does not contain patient information, please contact the > sender and properly > dispose of the e-mail. > > > >

1 0

Poor subcortical segmentation
by Mihaela Stefan 26 Nov '14

26 Nov '14

Hello Freesurfers, We noticed that our subcortical segmentations have a poor quality (see attachment). Is there a way to improve the segmentation in an automated fashion (avoiding long hours of manual editing)? We use FS 5.3 on Ubuntu 12.04. Thanks! Mihaela

2 5

kvlQuantifyHippocampalSubfieldSegmentations.sh resultsDirectory 1, 10 gives error line 22: cd: left: No such file or directory
by Knut J Bjuland 25 Nov '14

25 Nov '14

Dear Freesurfer experts, I have processed my subject with recon-all and subfields hippocampal segmentation. . I have tried to use kvlQuantifyHippocampalSubfieldSegmentations.sh on a subset of my subjects. I have these subjects A B C D E fsaverage in a folder. When I try to exclude fsaverage by using this command kvlQuantifyHippocampalSubfieldSegmentations.sh ~/subjects/hippo 1 5. I get this error message. cd A/ Quantifying subject A/ left Doing left side cd left /usr/local/freesurfer/bin/kvlQuantifyHippocampalSubfieldSegmentations.sh: line 22: cd: left: No such file or directory failed to do cd left I have issued this command both in tsch and in bash. However, when I run the command without any options in the same folder, it works but bails out as expected when it reaches fsaverage- Best regards, Knut J. Bjuland PhD candidate NTNU

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer November 2014