Freesurfer

freesurfer@nmr.mgh.harvard.edu

17 participants
30086 discussions

Spatial smoothing in Matlab
by Reza Rajimehr 18 Jul '14

18 Jul '14

Hi, For a simulation purpose, I want to load a surface into Matlab, assign a random value (taken from a normal distribution) to each vertex, do spatial smoothing on the surface with a given FWHM (e.g. 5mm), then do some computations with the smoothed values. I am wondering if there is a Matlab code that I can use for surface smoothing, similar to what FS-FAST preprocessing does in Freesurfer version 5 and above. Thanks, Reza ***************** Reza Rajimehr, MD Postdoctoral Research Fellow McGovern Institute for Brain Research Massachusetts Institute of Technology (MIT) NMR Martinos Center Massachusetts General Hospital (MGH) Harvard University *****************

3 2

retinotopic mapping using selxavg3-sess
by Alan Lee 18 Jul '14

18 Jul '14

Hello freesurferers, I ran this command (33333 is my subject name): selxavg3-sess -a Retinotopy.33333.lh -s 33333 and got this error message: # ---------- Using FSL's BET to Extract Brain------------------ # /usr/local/freesurfer/subjects/Retinotopy/33333/Retinotopy bet.fsl /tmp/mkbrainmask_14781/in.nii /tmp/mkbrainmask_14781/brain -m -f 0.1 /usr/lib/fsl/5.0/bin/bet2: error while loading shared libraries: libfslvtkio.so: cannot open shared object file: No such file or directory It seems that something is wrong with the file libfslvtkio.so, or perhaps some path problem (I've manually set $FSLDIR to /usr/lib/fsl/5.0/). Any feedback would be greatly appreciated. Below are the versions of software I'm using: Linux debian 3.2.0-4-486 #1 Debian 3.2.57-3+deb7u1 i686 GNU/Linux (run in virtualbox with a Windows 8 Host) FSL 5.0.6 freesurfer-Linux-centos4-stable-pub-v5.3.0 Many thanks, Alan -- Alan Lee sites.google.com/site/alanlflee/

3 5

Re: [Freesurfer] Spatial smoothing in Matlab
by Reza Rajimehr 18 Jul '14

18 Jul '14

Thanks Doug! I need to do this procedure 100 times, so I have to automate it. Do you have an example C-shell script for this? Or is there a way to call mri_surf2surf inside a "for loop" of Matlab? Thanks, Reza not that I know of. When I have to do this, I save the data to a file, then call mri_surf2surf with a system call, then load the result back into matlab.

1 0

Re: [Freesurfer] mri_convert
by Douglas Greve 18 Jul '14

18 Jul '14

You can use mri_convert to change the orientation in the header. If you run mri_convert without options it will give you lots of documentation doug On 7/18/14 9:17 AM, Fan, Qiuyun wrote: > Hi Doug, > > I think I figured it out. It's the angulation of the image plane. ImageOreintation (should have been axial slices) reads "0.99995101907245\0.00923454555106\-0.0035612672513\-0.0089978930043\0.69829945291611\-0.7157491962821", and was thus identified as coronal. So, nothing went wrong with mri_convert. > > As a follow up question, although technically, there's nothing wrong with it, it does look confusing intuitively. I'm wondering if there's a way to fool the nifti to label it as LPS at all? > > Thanks a lot! > > Best, > > Qiuyun > ________________________________________ > From: Fan, Qiuyun > Sent: Thursday, July 03, 2014 4:19 PM > To: Douglas N Greve > Subject: mri_convert > > Hi, Doug, > > I found an error when using mri_convert to convert image from dicom to nifti. I have one odd subject, whose orientation labels were wrong. The dicom-info-dump looks the same as the other subjects, and the images look fine too. > > If you have a chance to take a look at the images, here's the path to the "odd" subject:/autofs/cluster/hcps/data_preproc/HCPS_017/dcm/diff_1k/ > > An example of a "good" subject can be found here: /autofs/cluster/hcps/data_preproc/HCPS_015/dcm/diff_1k/ > > BTW, I'm using /usr/local/freesurfer/stable5_3_0. > > Please let me know if you need more information from me. > > Thanks a lot! > > Best, > > Qiuyun > >

1 0

Errors in Mapping of Brodmann Areas
by pfannmoelj＠uni-greifswald.de 18 Jul '14

18 Jul '14

Dear Experts, I am working with freesurfer 5.3 on Mac OSX. I applied recon-all to a data set. This has been done with the same data on a Linux system with Freesurfer 5.3 without any problems. Everything works fine on the Mac except of the labelling of Brodmann Areas. On the Mac the regions are not created as a single areal with a border at the outline. Instead they consist of an outside border and several small regions inside the label. I already tried to create the Brodmann Areas by myself using the command: mri_label2label --srclabel /.../subjects/fsaverage/label/lh.BA3b.label --srcsubject fsaverage --trglabel lh.BA3b.test.label --trgsubject subjid --regmethod surface --hemi lh This does not change the result. I attached a screen shot of the label in fsaverage and one of my subjects. Any help is highly appreciated. Regards pfannmoe

2 2

mni152reg error
by JacobML＠nmr.mgh.harvard.edu 18 Jul '14

18 Jul '14

Hello Freesurfer Experts, I am trying to view stats generated with FsFast in fslview/MNIspace, and in doing so I have run mni152reg to generate the registration matrix. That said, mni152reg did not complete correctly and below I have pasted the terminal output: Thu Jul 17 09:58:10 EDT 2014 setenv SUBJECTS_DIR /cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons cd /autofs/cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons /usr/local/freesurfer/stable5_3_0/bin/mni152reg --s CD_cerv_pat4_sess1_recon Linux purkinje 2.6.32-431.20.3.el6.x86_64 #1 SMP Thu Jun 19 21:14:45 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux fslregister --mov /usr/pubsw/packages/fsl/current/data/standard/MNI152_T1_2mm_brain.nii.gz --s CD_cerv_pat4_sess1_recon --reg /cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons/CD_cerv_pat4_sess1_recon/mri/transforms/reg.mni152.2mm.dat --dof 12 mkdir: cannot create directory `/usr/pubsw/packages/fsl/current/data/standard/tmp.fslregister.20833': Permission denied Log file is /cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons/CD_cerv_pat4_sess1_recon/mri/transforms/reg.mni152.2mm.dat.fslregister.log Thu Jul 17 09:58:11 EDT 2014 --mov /usr/pubsw/packages/fsl/current/data/standard/MNI152_T1_2mm_brain.nii.gz --s CD_cerv_pat4_sess1_recon --reg /cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons/CD_cerv_pat4_sess1_recon/mri/transforms/reg.mni152.2mm.dat --dof 12 $Id: fslregister,v 1.34.2.2 2012/11/08 16:18:56 greve Exp $ purkinje Linux purkinje 2.6.32-431.20.3.el6.x86_64 #1 SMP Thu Jun 19 21:14:45 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux nIters 1 -------------------------------------- /autofs/cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons mri_convert /cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons/CD_cerv_pat4_sess1_recon/mri/brainmask.mgz /usr/pubsw/packages/fsl/current/data/standard/tmp.fslregister.20833/refvol.fslregister.nii mri_convert /cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons/CD_cerv_pat4_sess1_recon/mri/brainmask.mgz /usr/pubsw/packages/fsl/current/data/standard/tmp.fslregister.20833/refvol.fslregister.nii niiWrite(): error opening file /usr/pubsw/packages/fsl/current/data/standard/tmp.fslregister.20833/refvol.fslregister.nii $Id: mri_convert.c,v 1.179.2.7 2012/09/05 21:55:16 mreuter Exp $ reading from /cluster/ablood/1/AdjustedParadigmFiles/Adjusted_recons/CD_cerv_pat4_sess1_recon/mri/brainmask.mgz... TR=2530.00, TE=3.25, TI=1100.00, flip angle=7.00 i_ras = (-1, 0, 0) j_ras = (-0, -0, -1) k_ras = (0, 1, 0) writing to /usr/pubsw/packages/fsl/current/data/standard/tmp.fslregister.20833/refvol.fslregister.nii... ERROR: failure writing /usr/pubsw/packages/fsl/current/data/standard/tmp.fslregister.20833/refvol.fslregister.nii I tried to run the fslregister command from within the subject's directory (in hopes to possible correct what may be a permissions issue), but this produces the same final error, "ERROR: failure writing /usr/pubsw/packages/fsl/current/data/standard/tmp.fslregister.24919/refvol.fslregister.nii." Lastly, probably to no surprise, the /.../mri/transforms/reg.mni152.2mm.dat file was never generated. Any ideas how I may be able to correct these errors? Thank you for the help! Jacob

3 4

selxavg3-sess error in ==> fast_selxavg3 at 434
by pfannmoelj＠uni-greifswald.de 18 Jul '14

18 Jul '14

Dear Experts, during an fsfast analysis an error occured after selxavg3-sess was executed. The corresponding commands were: mkanalysis-sess -fsd bold -surface self lh -fwhm 0 -event-related -paradigm sensory.par -nconditions 1 -spmhrf 0 -TR 2 -refeventdur 10 -nskip 8 -polyfit 2 -analysis analysis.lh -delay 0 -per-run -force -mcextreg mkcontrast-sess -analysis analysis.lh -contrast stim-v-base -a 1 selxavg3-sess -s ./sessdir -analysis analysis.lh -no-preproc -force The output of selxavg3-sess was: >> >> >> >> >> >> >> sxa3pwd = /.../evaluation >> sxa3cmd = /usr/local/freesurfer/fsfast/bin/selxavg3-sess -s ./sessdir -analysis analysis.lh -no-preproc -force >> /usr/local/freesurfer/fsfast/toolbox/fast_selxavg3.m >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> #@# sessdir ############################### /.../evaluation/sessdir ------------------------- $Id: fast_selxavg3.m,v 1.100.2.2 2012/11/30 18:40:38 greve Exp $ /usr/local/freesurfer/fsfast/toolbox/fast_selxavg3.m /usr/local/freesurfer/fsfast/toolbox/fast_ldanaflac.m /usr/local/freesurfer/matlab/MRIread.m ------------------------- outtop = /.../evaluation Extension format = nii.gz 1 stim-v-base.mat Excluding 8 points nruns = 1 autostimdur = outanadir = /.../evaluation/sessdir/bold/analysis.lh Excluding 8 points Found 145573/155188 (93.8) voxels in mask Creating Design Matrix ... creation time = 0.008 sec DoMCFit = 1 ntptot = 108, nX = 15, DOF = 93 Saving X matrix to /.../evaluation/sessdir/bold/analysis.lh/Xtmp.mat XCond = 64.7067 (normalized) Computing compensation for resdual AR1 bias 1 -0.5 -0.468819 (t=0.005249) 2 -0.25 -0.252194 (t=0.007053) 3 0 -0.0449419 (t=0.008126) 4 0.25 0.149516 (t=0.00935) 5 0.5 0.319324 (t=0.010795) AR1 Correction M: 0.0749452 1.26122 Computing contrast matrices OLS Beta Pass run 1 t= 0.0 Global Mean Global In-Mask Mean = (0.664951) ??? Error using ==> mrdivide Matrix dimensions must agree. Error in ==> fast_selxavg3 at 434 RescaleFactor = flac0.inorm/gmean; >> ------------------------------------------ ERROR: fast_selxavg3() failed\n Do you have an idea what went wrong? Sincerely yours pfannmoe

2 2

Re: [Freesurfer] unit of each index (thickness, volume, sulc, curv) after mri_glmfit-sim
by wang kangcheng 18 Jul '14

18 Jul '14

Thanks doug I first merge the data using command of mris_preproc and then do the GLM analysis using command of mri_glmfit. After that, using the following command to look the result: mri_surfcluster --in leg.volume.lh.10/g2v2/sig.mgh --csd /usr/local/freesurfer/average/mult-comp-cor/fsaverage/lh/cortex/fwhm15/pos/th13/mc-z.csd --mask leg.volume.lh.10/mask.mgh --cwsig leg.volume.lh.10/g2v2/cache.th13.pos.sig.cluster.mgh --vwsig leg.volume.lh.10/g2v2/cache.th13.pos.sig.voxel.mgh --sum leg.volume.lh.10/g2v2/cache.th13.pos.sig.cluster.summary --ocn leg.volume.lh.10/g2v2/cache.th13.pos.sig.ocn.mgh --oannot leg.volume.lh.10/g2v2/cache.th13.pos.sig.ocn.annot --annot aparc --csdpdf leg.volume.lh.10/g2v2/cache.th13.pos.pdf.dat --cwpvalthresh 1 --o leg.volume.lh.10/g2v2/cache.th13.pos.sig.masked.mgh --no-fixmni --bonferroni 2 --surf white Did you mean that I multiple the value by the number of vertices (NVtxs, which in the file of cache.th13.pos.sig.cluster.summary ) to get the total volume for the cluster? # ClusterNo Max VtxMax Size(mm^2) MNIX MNIY MNIZ CWP CWPLow CWPHi NVtxs Annot 1 4.623 7401 1647.03 -8.8 -49.3 5.4 0.00719 0.00559 0.00878 4199 isthmuscingulate 2 3.984 46872 1185.43 -7.7 -71.9 50.4 0.06375 0.05929 0.06820 2487 precuneus 3 3.960 57039 3597.23 -10.6 64.4 -5.0 0.00020 0.00000 0.00040 5545 frontalpole Wang Kangcheng Faculty of Psychology, Southwest University, China Email: kangchengwang0808(a)gmail.com / wangkangcheng.001(a)163.com Address: No.2 Tiansheng Road, BeiBei District, Chongqing, 400715,P.R.China

1 0

Re: [Freesurfer] unit of each index (thickness, volume, sulc, curv) after mri_glmfit-sim
by wang kangcheng 18 Jul '14

18 Jul '14

Thank you doug Data of thickness: 1.83629 3.01648 2.16078 corresponds to lateraloccipital, frontalpole and superiorparietal. Why the number of volume (e.g 1.39463 1.50616 2.53250) was also so small ? was the unit cm^3? Wang Kangcheng

2 1

unit of each index (thickness, volume, sulc, curv) after mri_glmfit-sim
by wang kangcheng 17 Jul '14

17 Jul '14

Dear freesurfer experts After using the command of mri_glmfit-sim (mri_glmfit-sim --glmdir aaa.thickness(or thickness, volume, sulc, curv ).lh.10 --cache 1.3 pos --cwp 1 --2spaces). I check the data in the fie of cache.th13.pos.y.ocn.dat and found that the data of volume was so small ( e.g 1.39463 1.50616 2.53250).Data of thickness was same as volume (e.g 1.83629 3.01648 2.16078). Could you tell the unit of each index? And could the data be used as further analysis, e.g correlation with IQ or other behaviors? Thank you Wang kangcheng

2 1

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