Freesurfer

freesurfer@nmr.mgh.harvard.edu

11 participants
30080 discussions

Freesurfer Citation in Paper
by Pedro Paulo Oliveira Jr 28 Feb '08

28 Feb '08

Is correct the statement below? FreeSurfer implementation of cortical surface-based analysis is distinct from other surface-based algorithms and conventional volume based morphometry with atlas mapping, as it does not necessarily depend on smoothing or matching to an atlas. Thanks in advance, Pedro Paulo Jr

3 3

Error loading lh.curv
by SHRIKANTH KULASHEKHAR 27 Feb '08

27 Feb '08

Hi, When I try loading curvature files on to inflated brains of the same subject, Freesurfer closes abruptly with the following error. % alloc: invalid block: 0xaec980: 0 0 68 Abort I am using Freesufer version stable4 on an Linux workstaion running on an AMD64 version of Debian 4.0, with kernel 2.6.24.2. Could anyone advice on how I could proceed ? Thanks in advance. Shriks 5, 50, 500, 5000 - Store N number of mails in your inbox. Go to http://help.yahoo.com/l/in/yahoo/mail/yahoomail/tools/tools-08.html

2 1

Subcortical calabel errors
by Marlisa Isom 27 Feb '08

27 Feb '08

Hello All, To begin I'm running 4.0.1 on Mac OSX Tiger (Intel). I'm having trouble with one subject in particular failing before generating the aseg.auto.mgz or aseg.mgz. I don't see anything amiss in the previously generated files and images, however, this subject has extremely large ventricles, which I think may be the issue. I get the same error whether running autorecon2, recon-all -calabel, etc. Here's the last some-odd lines of the log file: gca peak Unknown = 0.94091 ( 0) gca peak Left_Inf_Lat_Vent = 0.15806 (38) gca peak Left_Thalamus_Proper = 0.10842 (113) gca peak CSF = 0.21568 (39) gca peak Left_Accumbens_area = 0.39370 (75) gca peak Left_undetermined = 1.00000 (34) gca peak Left_vessel = 0.36599 (60) gca peak Left_choroid_plexus = 0.07952 (51) gca peak Right_Inf_Lat_Vent = 0.21306 (34) gca peak Right_Thalamus_Proper = 0.09033 (102) gca peak Right_Caudate = 0.20951 (79) gca peak Right_Hippocampus = 0.24233 (70) gca peak Right_Accumbens_area = 0.32477 (77) gca peak Right_vessel = 0.58618 (58) gca peak Right_choroid_plexus = 0.14637 (47) gca peak Fifth_Ventricle = 0.96184 (31) gca peak WM_hypointensities = 0.12793 (82) gca peak non_WM_hypointensities = 0.07548 (80) gca peak Optic_Chiasm = 0.48149 (75) not using caudate to estimate GM means setting label Left_Hippocampus based on Right_Hippocampus = 1.00 x + 0 setting label Left_Caudate based on Right_Caudate = 1.00 x + 0 estimating mean gm scale to be 0.99 x + 0.0 estimating mean wm scale to be 1.00 x + 0.0 estimating mean csf scale to be 0.93 x + 0.0 46135 gm and wm labels changed (%25 to gray, %75 to white out of all changed labels) 401 hippocampal voxels changed. 0 amygdala voxels changed. pass 1: 101186 changed. image ll: -2.438, PF=1.000 pass 2: 16699 changed. image ll: -2.435, PF=1.000 pass 3: 6379 changed. pass 4: 2882 changed. Segmentation fault Darwin hlcpro2.psychology.wustl.edu 8.11.1 Darwin Kernel Version 8.11.1: Wed Oct 10 18:23:28 PDT 2007; root:xnu-792.25.20~1/RELEASE_I386 i386 i386 recon-all exited with ERRORS at Thu Feb 21 23:29:10 CST 2008 I've seen a -nobigventricles flag, is there perhaps a -bigventricles option? Any suggestions are appreciated. Thanks! -Marlisa Marlisa Isom Research Coordinator Head Research Laboratory Washington University in St. Louis

3 3

Talairach Failure Detection
by wanlin 27 Feb '08

27 Feb '08

- Hi, we met some errors as below when we run "recon-all -subjid SUBNAME -autorecon1". Similar questions have been asked in http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg06452.html http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg06739.html But the answer seems not useful for our situation. We are performing data processing pipeline with 3T MRI volume image. If we add -notal-check, the errors will disappear but we are not sure whether it's ok without Talairach check. Can anyone help us how to deal with them when such errors appeared? thanks #-------------------------------------------- #@# Talairach Failure Detection Wed Feb 27 07:43:35 EST 2008 /Data/MAS/FreeSurfer/CPU7/3914_20070206_3dt1/mri talairach_afd -T 0.005 -xfm transforms/talairach.xfm talairach_afd: Talairach Transform: transforms/talairach.xfm OK (p=0.7627, pval=0.6675 >= threshold=0.0050) awk -f /usr/local/freesurfer4.0.2/bin/extract_talairach_avi_QA.awk /Data/MAS/FreeSurfer/CPU7/3914_20070206_3dt1/mri/transforms/talairach_avi.log TalAviQA: 0.94477 z-score: -7 ERROR: Talairach QA check failed! z-score = -7 < -5 = threshold Linux thalamus2007 2.6.23.15-80.fc7 #1 SMP Sun Feb 10 16:52:18 EST 2008 x86_64 x86_64 x86_64 GNU/Linux recon-all exited with ERRORS at Wed Feb 27 07:43:35 EST 2008

2 1

qdec
by Feng-Xian Yan 27 Feb '08

27 Feb '08

hi, I have a problem. I analysis my data using qdec. But, after runing the command (recon-all -s subject1 -qcache) I can't find the qdec.table.dat and the discrete factor files. The following commands that I type to run. First, I set SUBJECTS_DIR to a directory. setenv SUBJECTS_DIR $SUBJECTS_DIR/group_analysis cd $SUBJECTS_DIR Then, I run the pre-smoothed fsaverage surfaces recon-all -s subjects1 -qcache Could you tell me what's wrong? Thank you~~

2 1

Re: [Freesurfer] autorecon1 errors
by Jenni Pacheco 26 Feb '08

26 Feb '08

Hi Mike, I've run into the same problem you see in number 1 - the SPLINE_CUBIC_SET - Fatal Error!, with a fine talaraich. I was able to fix this by using the -no1d flag in mri_normalize. You can add this as an xopt to that subject so that it is always run: http://surfer.nmr.mgh.harvard.edu/fswiki/recon_2dall#head-25c704d7d03b1c22e… I think this occurs when there is a bigger bias field in left-right direction than in the anterior-posterior direction (although, maybe it's I/S, I forget now). I'll leave the other two problems for someone more qualified! Jenni -- Learning and Memory Lab The University of Texas at Austin Department of Psychology 1 University Station A8000 Austin, TX 78712 phone: (512) 471-2727 fax: (512) 471-5935 email: jpacheco(a)mail.utexas.edu > On Tue, Feb 26, 2008 at 3:29 PM, Michael Harms <mharms(a)conte.wustl.edu> > wrote: > > > > > Hello, > > We recently just ran 80 elderly brains through -autorecon1 using FS > > v4.0.2, and encountered 3 different types of errors that we are not > > entirely sure how to handle: > > > > (1) During mri_normalize of 1 subject we encountered: > > SPLINE_CUBIC_SET - Fatal error! > > The number of data points N must be at least 2. > > The input value is 1. > > > > The FS archive indicates to check the talairach transform given this > > error, but the talairach.xfm appears fine for this subject (as assessed > > using tkregister2). > > > > (2) During mri_watershed of 1 subject: > > Sorting... > > mri_watershed Error: > > Problem in the COG calculation > > > > This subject had a slightly warped talaraich.xfm, but I think more > > relevantly, its talairach_with_skull.lta was way off. > > > > (3) During mri_watershed of 5 subjects: > > mri_watershed Error: > > GLOBAL region of the brain empty ! > > > > For all 5 of these subjects, the talairch_with_skull.lta also had > > obvious problems. > > > > RE (1): Any insight would be appreciated. > > > > RE (2) and (3): I suspect that we should try running with the -no- > > wsgcaatlas flag, so that talairach_with_skull.lta is longer used by > > mri_watershed. Should we then substitute the -wsatlas flag instead? > > (Its not clear to me which atlas is used when that particular flag is > > included). > > > > Last, I thought you might be interested in these errors, because they > > directly relate to my recent question regarding the robustness of > > talairach_with_skull.lta. At least during the -autorecon1 (i.e., > > initial) creation of this lta file, it appears that there is some > > possibility of a bad talairach_with_skull registration (6 of 80 in this > > particular subject set). > > > > thanks, > > Mike H. > > -- > > Michael Harms, Ph.D. > > -------------------------------------------------------------------- > > Conte Center for the Neuroscience of Mental Disorders > > Washington University School of Medicine > > Department of Psychiatry, Box 8134 > > Renard Hospital, Room 6615 Tel: 314-747-6173 > > 660 South Euclid Ave. Fax: 314-747-2182 > > St. Louis, MO 63110 Email: mharms(a)wustl.edu > > -------------------------------------------------------------------- > > > > > > _______________________________________________ > > Freesurfer mailing list > > Freesurfer(a)nmr.mgh.harvard.edu > > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > >

1 0

autorecon1 errors
by Michael Harms 26 Feb '08

26 Feb '08

Hello, We recently just ran 80 elderly brains through -autorecon1 using FS v4.0.2, and encountered 3 different types of errors that we are not entirely sure how to handle: (1) During mri_normalize of 1 subject we encountered: SPLINE_CUBIC_SET - Fatal error! The number of data points N must be at least 2. The input value is 1. The FS archive indicates to check the talairach transform given this error, but the talairach.xfm appears fine for this subject (as assessed using tkregister2). (2) During mri_watershed of 1 subject: Sorting... mri_watershed Error: Problem in the COG calculation This subject had a slightly warped talaraich.xfm, but I think more relevantly, its talairach_with_skull.lta was way off. (3) During mri_watershed of 5 subjects: mri_watershed Error: GLOBAL region of the brain empty ! For all 5 of these subjects, the talairch_with_skull.lta also had obvious problems. RE (1): Any insight would be appreciated. RE (2) and (3): I suspect that we should try running with the -no- wsgcaatlas flag, so that talairach_with_skull.lta is longer used by mri_watershed. Should we then substitute the -wsatlas flag instead? (Its not clear to me which atlas is used when that particular flag is included). Last, I thought you might be interested in these errors, because they directly relate to my recent question regarding the robustness of talairach_with_skull.lta. At least during the -autorecon1 (i.e., initial) creation of this lta file, it appears that there is some possibility of a bad talairach_with_skull registration (6 of 80 in this particular subject set). thanks, Mike H. -- Michael Harms, Ph.D. -------------------------------------------------------------------- Conte Center for the Neuroscience of Mental Disorders Washington University School of Medicine Department of Psychiatry, Box 8134 Renard Hospital, Room 6615 Tel: 314-747-6173 660 South Euclid Ave. Fax: 314-747-2182 St. Louis, MO 63110 Email: mharms(a)wustl.edu --------------------------------------------------------------------

1 0

Calling mri_convert from Java code
by Alexander Zaitsev (CNI) 26 Feb '08

26 Feb '08

Hello: I need to call mri_convert utility from Java code. The problem is when I am making a system call, I am getting the following error : >>>>> FreeSurfer license file /Application/freesurfer/.license not found. If you are outside the NMR-Martinos center, go to http://surfer.nmr.mgh.harvard.edu to get a valid license file (it's free). If you are inside the NMR-Martinos center, make sure to source the standard environment. <<<<<< The mri_convert runs from the shell without problem and the .license file is in the right directory. I am including the following env variables into the system call: >>> String[] env = {"FREESURFER_HOME=/Application/freesurfer", "FSFAST_HOME=/Applications/freesurfer/fsfast", "SUBJECTS_DIR=/Applications/freesurfer/subjects", "MNI_DIR=/Applications/freesurfer/mni", "GSL_DIR=/Applications/freesurfer/lib/gsl", "FSL_DIR=/Applications/freesurfer/fsl", "FUNCTIONALS_DIR=$FREESURFER_HOME/sessions", "FSLOUTPUTTYPE=NIFTI_GZ", sPATH, "DISPLAY=:0.0", "HOME=/tmp", "DYLD_LIBRARY_PATH=/Applications/freesurfer/lib/gsl/lib", "MINC_BIN_DIR=/Applications/freesurfer/mni/bin", "MINC_LIB_DIR=/Applications/freesurfer/mni/lib", "MNI_DATAPATH=/Applications/freesurfer/mni/data", "LOCAL_DIR=/Applications/freesurfer/local", "FMRI_ANALYSIS_DIR=/Applications/freesurfer/fsfast", "MNI_PERL5LIB=/Applications/freesurfer/mni/lib/../System/Library/Perl/5.8.6", "PERL5LIB=/Applications/freesurfer/mni/lib/../System/Library/Perl/5.8.6", "LD_LIBRARY_PATH=/Applications/freesurfer/lib/gsl/lib", "FSLDIR=/Applications/freesurfer/fsl", "FSL_BIN=/Applications/freesurfer/fsl/bin", "FSLTCLSH=/Applications/freesurfer/fsl/bin/tclsh", "FSLWISH=/Applications/freesurfer/fsl/bin/wish", "FSLGNUPLOT=/Applications/freesurfer/fsl/bin/gnuplot", "FSLDISPLAY=/Applications/freesurfer/fsl/bin/display", "FSLCONVERT=/Applications/freesurfer/fsl/bin/convert", "FSLBROWSER=/Applications/freesurfer/fsl/tcl/fslwebbrowser", "FSLCONFDIR=/Applications/freesurfer/fsl/config" }; <<< Am I missing something? Yes, the mri_convert runs fine when called from the shell. And I don't see any other env variables related to the free surfer. Thank you, Alex

2 1

inflation and registration to a sphere
by Mishkin Derakhshan 26 Feb '08

26 Feb '08

Hi, After reading the wiki and the references on the wiki, specifically [1] and [2], I'm still a little confused about how the inflation and registration to a sphere occurs. ie. what energy functions are being minimized at each step. I think it is a 3 step process, where 1. Surfaces are inflated by minimizing the energy function Js. This creates ?h.inflated. Js = 1/2V (EE ||xi-xn||^2) + lambda-d*Jd (8) in [1] 2. The inflated surfaces are then projected to a sphere by minimizing the energy function J. This creates ?h.sphere J = ??? 3. The sphere is registered to the average by minimizing the energy function J. This creates ?h.sphere.reg. J = Jp + lambda-d*Jd + lambda-a*Ja (5) in [2] Is there a step I am missing? Are my energy functions correct? thanks, mishkin [1] Cortical Surface-Based Analysis II: Inflation, Flattening, and a Surface-Based Coordinate System, Fischl, B., Sereno, M.I., Dale, A.M., (1999). NeuroImage, 9(2):195-207. [2] High-resolution inter-subject averaging and a coordinate system for the cortical surface, Fischl, B., Sereno, M.I., Tootell, R.B.H., and Dale, A.M., (1999). Human Brain Mapping, 8:272-284(1999).

2 1

Editing annotations
by Pierson, Ronald K 26 Feb '08

26 Feb '08

Forgive me if this has been covered before, but I was unable to find an answer after searching for a bit. I have been considering editing the final surface parcellation for those scans that have problems in certain areas. Can these corrections then be used to correct the surface registration for the input into Qdec? Also, to test out the editing process I have been simply importing the annot file, editing the labels, then exporting it. Is this the correct thing to do, and is there a tutorial anywhere about this kind of editing? Thanks, Ronald Pierson

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