Freesurfer November 2005

freesurfer@nmr.mgh.harvard.edu

49 participants
73 discussions

Question
by Wan Park 26 Sep '06

26 Sep '06

Hello, my name is Wan Park and I am working at Columbia University. After converting General Electric DIACOM files into COR, I go to the Setup Structural Scans. There I set the Structural Scan Dir and then Convert/Average. After that, Start the Process Volume, but it seems that there is a problem when it tries to strip the skull. i receive this message: Images (brain): /Users/../Volumes/ste/freesurfer/subjects/scott/mri/brain/COR-001.. not found ... strip skull first So I go to see the logs and I find this: /Users/wanpark/Desktop/FreeSurfer/freesurfer/bin/Darwin/mri_normalize /Users/../Volumes/stem/freesurfer/subjects/scott/mri/orig /Users/../Volumes/stem/freesurfer/subjects/scott/mri/T1 reading from /Users/../Volumes/stem/freesurfer/subjects/scott/mri/orig... normalizing image... 3d normalization pass 1 of 2 building Voronoi diagram... performing soap bubble smoothing... 3d normalization pass 2 of 2 building Voronoi diagram... performing soap bubble smoothing... writing output to /Users/../Volumes/stem/freesurfer/subjects/scott/mri/T1 3D bias adjustment took 3 minutes and 23 seconds. Command Finished /Users/wanpark/Desktop/FreeSurfer/freesurfer/bin/Darwin/mri_watershed /Users/../Volumes/stem/freesurfer/subjects/scott/mri/T1 /Users/../Volumes/stem/freesurfer/subjects/scott/mri/brain ************************************************************ The input file is /Users/../Volumes/stem/freesurfer/subjects/scott/mri/T1 The output file is /Users/../Volumes/stem/freesurfer/subjects/scott/mri/brain If this is incorrect, please exit quickly the program (Ctl-C) *************************WATERSHED************************** preflooding height equal to 25 percent Sorting... first estimation of the COG coord: x=125 y=148 z=132 r=54 first estimation of the main basin volume: 693155 voxels ******************************************************** ******************************************************** ******************************************************** White matter intensity 0 is lower than CSF intensity 26. Please examine input images. Will terminate ... ******************************************************** ******************************************************** ******************************************************** ************************************************************** MRIstipSkull failed. ************************************************************** Command Finished Could you tell me what's wrong? Thanks, Wan Park

4 3

retinotopy and FsFast
by Jane Aspell 17 May '06

17 May '06

Hi Thanks for all your previous help with freesurfer. I am trying to use FsFast to do retinotopy analysis and have been following the instructions at: https://surfer.nmr.mgh.harvard.edu/fswiki/FsFastIndividualRetinotopyAnalysis I come undone at the stage of running the actual analysis, where the command should be: >sfa-sess -s avdk -a rtopy what does 'avdk' refer to? when i run it i get: ERROR: could not find session avdk when i try it with the arguments: >sfa-sess -analysis rtopy -sf sessid which is similar to the arguments needed for 'selxavg-sess' i get a similar error of not been able to find my (particular subject's) session. i believe i have set up the all the directory structures etc correctly for the sessions format as previous steps worked fine. any ideas what's going wrong? thanks Jane Aspell

5 18

Short Course in fMRI
by Doug Greve 10 Apr '06

10 Apr '06

Mark Vangel and I are offering a short course in the statistical signal processing of fMRI. October 31 (Mon) & November 1 (Tue), 2005 7th Floor Auditorium Building 149 Charlestown Navy Yard Free, but limited to 40 attendees. Register at: http://www.nmr.mgh.harvard.edu/martinos/training/shortCoursefMRIStats.php doug

1 1

N3 in FreeSurfer
by Michele Perry 05 Dec '05

05 Dec '05

Is the normalization software called N3 a standard part of the dev release of Freesurfer? Michele

4 9

motion correction during recon-all
by Don Hagler 02 Dec '05

02 Dec '05

I want to average multiple structural scans for input into the surface recon. Two are axial and one is sagittal. (I plan to correct for the gradient inhomogeneities in the sagittal images, so that shouldn't be an issue.) Anyway, the problem comes when recon-all does "motion correction" to align the scans. For some strange reason, the axial and sagittal scan protocols use slightly different TRs and TEs (e.g. TR=7.996 vs. 8.012). For some even stranger reason, this brings recon-all to a screaching halt. I'm guessing that this is actually a MINC problem but I wonder if there is a way to make it ignore the TR and TE when aligning two structural scans. My work around is to explicitly set the TR and TE for the sagittal scan when running mri_convert: mri_convert -tr 8.012 -te 3.136 -it dicom i\*.1 $destdir/$mgzname.mgz

2 3

neuroimaging job announcement
by David Ziegler 30 Nov '05

30 Nov '05

The Behavioral Neuroscience Lab at MIT is seeking a technical assistant who will take a lead role in the design, implementation, and analysis of our structural and functional MRI studies. A candidate with Freesurfer/FSL experience will have a distinct advantage. Here is a link to the full job description and online application page: http://sh.webhire.com/servlet/av/jd?ai=631&ji=1642175&sn=I -- David A. Ziegler Department of Brain and Cognitive Sciences Massachusetts Institute of Technology Building 46-5121, 32 Vassar St Cambridge, MA 02139 Tel: 617-258-0765 Fax: 617-253-1504 daz(a)mit.edu <mailto:daz@mit.edu>

1 0

RE: [Freesurfer] version history question
by Goldman, Aaron (NIH/NIMH) 29 Nov '05

29 Nov '05

Okay. Related to that question, if I reran people with the latest tools, would it be possible/make sense to read in my control point files from the old version? Thanks, -Aaron- -----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Tuesday, November 29, 2005 4:17 PM To: Goldman, Aaron (NIH/NIMH) Cc: freesurfer(a)nmr.mgh.harvard.edu Subject: Re: [Freesurfer] version history question Hi Aaron, one of the things we are trying to put in place is something like recon-all -rerun that will retain all manual interventions but reapply the newest tool to all your data. And yes, those tools have changed in the last year so you don't want to compare across versions if you can help it. cheers, Bruce On Tue, 29 Nov 2005, Goldman, Aaron (NIH/NIMH) wrote: > Hi, I'm interested in analyzing subcortical segmentations, and have a question about potential systematic differences between freesurfer versions. Specifically, I have a large number of subjects who've been through a year-old version of recon-all -stage1 (the skull strip, normalization, etc), and was wondering if I need to keep using the old version for consistency's sake. Have those initial steps been substantially upgraded or changed over the past year? Also, is there somewhere I can find a guide to version history? > > > > The versions concerned are: > > OLD: 08/31/04, recon-all version 1.5, tosa > > NEW: 09/28/05, recon-all version 1.114, nicks > > > > Thanks, > > > > -Aaron- > > > > Aaron L. Goldman > > Post-Bacc IRTA > > Genes, Cognition and Psychosis Program, NIMH > > Building 10, Room 3C108 > > Phone: (301) 435-0944 > > Email: aarong(a)mail.nih.gov > > > >

2 1

version history question
by Goldman, Aaron (NIH/NIMH) 29 Nov '05

29 Nov '05

Hi, I'm interested in analyzing subcortical segmentations, and have a question about potential systematic differences between freesurfer versions. Specifically, I have a large number of subjects who've been through a year-old version of recon-all -stage1 (the skull strip, normalization, etc), and was wondering if I need to keep using the old version for consistency's sake. Have those initial steps been substantially upgraded or changed over the past year? Also, is there somewhere I can find a guide to version history? The versions concerned are: OLD: 08/31/04, recon-all version 1.5, tosa NEW: 09/28/05, recon-all version 1.114, nicks Thanks, -Aaron- Aaron L. Goldman Post-Bacc IRTA Genes, Cognition and Psychosis Program, NIMH Building 10, Room 3C108 Phone: (301) 435-0944 Email: aarong(a)mail.nih.gov

2 1

mri_ca_normalize error
by Michele Perry 28 Nov '05

28 Nov '05

Hi, mri_ca_normalize failed for a subject. Here is the CA normalize portion of recon-all.log: =========================================== CA Normalize Thu Nov 17 17:02:05 PST 2005 $SUBJECTS_DIR/<subjid>/mri mri_ca_normalize -mask brain.mgz nu.mgz /space/monkeys/1/pubsw/packages/freesurfer/RH4-x86_64/average/RB40_talairach_2005-07-20.gca transforms/talairach.lta norm.mgz using MR volume brain.mgz to mask input volume... reading 1 input volumes reading atlas from '/space/monkeys/1/pubsw/packages/freesurfer/RH4-x86_64/average/RB40_talairach_2005-07-20.gca'... setting gca type = Normal gca type Direction cosines read: x_r = -1.0000, y_r = 0.0000, z_r = 0.0000 x_a = 0.0000, y_a = 0.0000, z_a = 1.0000 x_s = 0.0000, y_s = -1.0000, z_s = 0.0000 c_r = 0.0000, c_a = 0.0000, c_s = 0.0000 reading transform from 'transforms/talairach.lta'... mghRead: encountered error executing: 'zcat $SUBJECTS_DIR/<subjid>/mri/nu.mgz',frame -1 mri_ca_normalize: could not read input MR volume from nu.mgz reading input volume from nu.mgz... ERROR: mri_ca_normalize with non-zero status ======================================================= $SUBJECTS_DIR/<subjid>/mri contains nu.mgz and the talairach looks normal. How do I fix this? Michele

3 2

tkregister2
by Margo McKenna Benoit 28 Nov '05

28 Nov '05

Hi, I am trying to manually check the talairach registration from recon-all using tkregister2, but it appears to be looking for files in <subjectname>talairach/mri/orig. This directory structure does not exist in my dataset. The tutorial recommended the following command line: *tkregister*2 --mgz --s <subjectname> --fstal --surf orig --fstal Check and edit the talairach registration that was created during the FreeSurfer reconstruction. Sets the movable volume to be $SUBJECTS_DIR/talairach/mri/orig and sets the registration file to be $SUBJECTS_DIR/subjectid/transforms/talairach.xfm. User must have write permission to this file. Do not specify --reg with this flag. It is ok to specify --regheader with this flag. The format of the anatomical can be changed with --mgh or --mghz My data is in /space/lucy/2/users/margo/subjects/avm002 and I am using the std environment. Can I run this command without the --fstal? What should I specify as the target and movable volumes in that case? Thank you, Margo

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 November 2005