Freesurfer August 2006

freesurfer@nmr.mgh.harvard.edu

48 participants
74 discussions

having problem in running recon-all...
by Shahabuddin Ansari 07 Aug '06

07 Aug '06

Hello, I am having problem in running the following command: recon-all -i 2996926_02.img -s 2996926_02 I used to run the command before without any problem. Recently my linux has been updated to suse version, would that be a source of problem. The console display the following error: INFO: SUBJECTS_DIR is /ensc/IMAGE/opt/FREESURFER/subjects Actual FREESURFER_HOME /ensc/IMAGE/opt/FREESURFER Linux ilpc5 2.6.13-15.10-bigsmp #1 SMP Fri May 12 16:36:41 UTC 2006 i686 i686 i386 GNU/Linux mri_convert: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_convert: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_normalize: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_watershed: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_segment: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_fill: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_tessellate: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mris_smooth: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mris_inflate: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mris_sphere: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mris_fix_topology: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mris_euler_number: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mris_make_surfaces: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mris_register: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mrisp_paint: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_label2label: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_em_register: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_ca_normalize: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_ca_register: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory mri_ca_label: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory /ensc/IMAGE/opt/FREESURFER/subjects/2996926_02 mri_convert /ensc/IMAGE4/PROJECTS/HD_AYLWARD/DATA/PATIENTDATA/SHAHAB/2996926_02/2996926_02.img /ensc/IMAGE/opt/FREESURFER/subjects/2996926_02/mri/orig/001.mgz mri_convert: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory Linux ilpc5 2.6.13-15.10-bigsmp #1 SMP Fri May 12 16:36:41 UTC 2006 i686 i686 i386 GNU/Linux recon-all exited with ERRORS at Sun Aug 6 19:08:34 PDT 2006 Shahab --------------------------------- Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail Beta.

2 1

Problem copying subject directories
by Henry Lütcke 05 Aug '06

05 Aug '06

Hi, I ran autorecon-all on a Suse10.0 computer and it finished fine. when I copy the respective subject directory to another computer (Suse10.1) I can no longer view the surfaces with tksurfer (only a very tiny part is displayed; see attachment). I can view the volume in tkmedit fine and the surface is displayed on the computer where the analysis has been run. any ideas what might be causing this problem? thanks a lot, Henry

2 2

Labels from segmentation
by Johannes Haushofer 03 Aug '06

03 Aug '06

Dear all, I'd like to create labels for certain structures based on their segmentation labels (e.g. a label containing all voxels in orbitofrontal cortex). Is there a way to automatize this? At the moment I'm using tkmedit to mark all voxels with the same segmentation label, then saving them as a label; that's a bit cumbersome... any hints? Many thanks! johannes

3 2

DICOM conversion
by Valentin Piech 03 Aug '06

03 Aug '06

Hello everyone, I ran into the same problem converting GE multi-volume data from DICOM to nifti 4D with mri_convert as Adam Thomas in the [Freesurfer] mri_convert: dicom to nifti 4D <https://mail.nmr.mgh.harvard.edu/pipermail//freesurfer/2006-July/003324.html> thread did: I got one volume containing several thousand slices instead of the 4D volume. I also downloaded the improved version that Doug posted in the thread, and it solved my problems until recently. I have a new data set that won't convert anymore, and "mri_convert -it dicom -ot nii -dicomread2 ./mydicomfile.DCM ./myniftifile.nii" exists with the following error: *** glibc detected *** malloc(): memory corruption 0x083ba568 *** The data set is taken with the exact same protocol, except that it contains more volumes (227 instead of 176 volumes with 29 slices each). According to Doug in this thread [Freesurfer] generic dicom converting <https://mail.nmr.mgh.harvard.edu/pipermail//freesurfer/2006-July/003477.html>, there is a new version out that may solve my problem, but as it is part of the DEV version, I cannot download it as far as I know since I am not at MGH. Is there any way to download it that I don't know of? Also, I am normally using the x86_64 version, while the one posted in the previous thread is only the x86 one, so I have to have the 32 bit version of freesurfer installed and use its path when I run this version of mri_convert. Thank you very much for your help, Valentin -- Valentin Piech, Graduate Fellow Gilbert Lab for Neurobiology & Reeke Lab for Biological Modelling Rockefeller University 1230 York Ave Box 193 New York, NY 10021 PH: (212) 327-7646

2 1

query regarding T2/T1
by Chacko Cherian 03 Aug '06

03 Aug '06

Hello all, Can we use tkmedit for T2-weighted images or is it restricted to T1 -weighted? Thanks Chacko.

2 1

importing high-res scans
by Marie Schaer 03 Aug '06

03 Aug '06

Hi everybody, I am trying to process high res monkey scans in FreeSurfer (voxel size 0.31 x 1 x 0.31 mm). For converting them I use: mri_convert rawdata.img orig.mgz -iis 0.31 -ijs 1 -iks 0.31 -cm But I actually get a very small image in tkmedit. Further, moving each slice in any of the three directions give a volume index change of 1 and voxel RAS change of 3 (i.e. on the coronal slice number 128, the volume index 128 128 128 is equivalent to the volume RAS 0 0 0; going to the coronal slice 129 without moving the cursor will give the volume index 128 128 131, and RAS 0 1 0). Since RAS are the coordinates in millimeter, I actually expected that any change of slice would be accompanied by a 0.31 change on RAS and variation of 1 on volume index... Is there any basic step that I missed? Thank you very much, Marie Schaer

2 3

error at autorecon2
by Chacko Cherian 02 Aug '06

02 Aug '06

Hello all, Here's an error I got while doing 'recon-all -autorecon2 -s <subject_name>. Could someone please clarify. Thank you. Chacko Cherian. #@# EM Registration Wed Aug 2 16:35:22 MST 2006 /home/ttest/freesurfer/subjects/case1/case1/case1/mri mri_em_register -mask brainmask.mgz nu.mgz /home/ttest/freesurfer/average/RB_all_2006-02-15.gca transforms/talairach.lta using MR volume brainmask.mgz to mask input volume... reading 1 input volumes... logging results to talairach.log reading '/home/ttest/freesurfer/average/RB_all_2006-02-15.gca'... setting gca type = Normal gca type average std = 7.4 using min determinant for regularization = 5.5 0 singular and 1662 ill-conditioned covariance matrices regularized reading 'nu.mgz'... Segmentation fault (core dumped) Linux ibmlnx2.samaritan.edu 2.4.20-24.9 #1 Mon Dec 1 11:35:51 EST 2003 i686 i686 i386 GNU/Linux recon-all exited with ERRORS at Wed Aug 2 16:35:41 MST 2006 __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

dicom format
by Chacko Cherian 02 Aug '06

02 Aug '06

I got this error while running a single subject in DICOM format.the 'recon-all -autorecon1 -subjid <subject_name> ' stopped at this stage. INFO: Modifying talairach volume c_(r,a,s) based on average_305 MRInormalize: could not find any valid peaks mri_normalize: normalization failed Linux ibmlnx2.samaritan.edu 2.4.20-24.9 #1 Mon Dec 1 11:35:51 EST 2003 i686 i686 i386 GNU/Linux recon-all exited with ERRORS at Wed Aug 2 14:16:22 MST 2006 __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

dicom files
by Chacko Cherian 02 Aug '06

02 Aug '06

Hello all, I have 180 slices of a subject in DICOM format.I am trying to run the 'recon-all' procedure for this subject. Are these slices seen by Freesurfer as individual acquisitions or as a single dicom file? In case they are percieved as individual acquisitions,do I need to manually type all the file names(for each slice) for the recon-all procedure or is there a simpler way. THanks in advance Chacko Cherian. Arizona State University Graduate Research Assistant __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

4 3

mri_watershed question
by Avram Holmes 02 Aug '06

02 Aug '06

All, I am interested in using mri_watershed to remove the skull from an inplane low res structural scan (not an MPRAGE). However, when I try to I get the following message followed by the program crashing. Mode: Atlas analysis ********************************************************* The input file is P1.mgz The output file is brainmask.auto.mgz If this is incorrect, please exit with CTL-C changing type of input volume to 8 bits/voxel... MRIchangeType: Building histogram *************************WATERSHED************************** preflooding height equal to 25 percent Sorting... first estimation of the COG coord: x=127 y=133 z=24 r=80 first estimation of the main basin volume: 2160181 voxelsSegmentation fault (core dumped) I am sure that I am missing some simple flag. Does anyone know a solution to this before I run off and start using BET. Thanks, Avram

2 1

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer August 2006