Freesurfer

freesurfer@nmr.mgh.harvard.edu

19 participants
30088 discussions

recon-all taking a really long time
by Jonathan McDaniel 19 Jul '11

19 Jul '11

Hi, I have been running a surface reconstruction for over 19 days now, and it is still in the process of correcting defects. Based on how long it took to do one hemisphere (it is now on the second) it appears it will be done hopefully within 3 days. I have been concurrently running other subjects on the same quadcore machine, using 3 of the 4 processors to perform the recons, and they have been finishing in a normal amount of time. Has anyone else had a brain take this long to run? Should I suspect that something is wrong with the inputted dicom files? Are manual edits necessary to fix some of the defects to speed up the process? Thanks for your answers! Jonathan

3 2

Having Matlab portion run as compiled matlab
by Bill Taylor 19 Jul '11

19 Jul '11

So I peaked in archive and did not get a hit. The situation is that the University of Wisconsin Madison site license prohibits us running Matlab in a cluster unless it is compiled. I am assisting a Neurology lab with having a Condor workflow which this time round will do both Matlab and FSL post processing. To compile the Matlab one needs to compile the entry points. But to run it, we'd have to change how it is called. Anyone have pointers for me? Bill Taylor Center for High Throughput Computing Condor Project Computer Sciences University of Wisconsin, Madison 608-219-4430

3 3

Running autorecon2-wm
by Antonella Kis 19 Jul '11

19 Jul '11

Hi Bruce, Based on your last suggestion, after running the recon-all on my NPI001 subject: recon-all -s NPI001 -i /usr/local/freesurfer/subjects/june2011/_orig_/NPI001.nii.gz -qcache -all and editing the wm, I saved all my edits (using the tool box menu as Save Volume As) under wm_edits_before.mgzm and I checked to make sure the edits are there/saved. Now I need to run autorecon2-wm and autorecon3 and I am not sure if in the subjid I still need to write NPI001 (in this case how Freesurfer knows that my wm edits are in the wm_edits_before) or I need to run the following syntax: recon-all -autorecon2-wm -autorecon3-s NPI001 -i /usr/local/freesurfer/subjects/june2011/_orig_/NPI001/mri/wm_edits_before.mgz ? Thank you very much. Antonella

3 2

which files for thickness and volume stats?
by Rebecca McCourt 18 Jul '11

18 Jul '11

Hi all, I'm wondering which files I should use to find volumetrics data... I need cortical thickness (whole-brain and each hemisphere), GM/WM volumes (cortical) and whole brain volume. I know about asegstats2table but I read that aseg isn't that reliable for WM/GM. What's the difference between mris_anatomical_stats and mri_segstats, and which of these should I use to get the data specified above? Thanks! Rebecca M University of Alberta

1 0

Re: [Freesurfer] Problem registering labels to native space
by John Richey 18 Jul '11

18 Jul '11

Hi Doug, this is indeed an fsfast analysis. I used the fsaverage subject included with the 5.1 distribution. Here is the code that I used to produce the registration matrix for one subject. Note that I did use mri_convert to change the .mgz files into .nii.gz The use of flirt (below) is more of an ad-hoc solution to register each subject's anatomical image to fsaverage. I'm not wedded to this method if there's a better alternative. bet ${SUBJECTS_DIR}/${SUBJECT}/mri/nu.nii.gz \ ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet.nii.gz \ -f 0.3 -g 0 flirt -in ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet.nii.gz \ -ref ${SUBJECTS_DIR}/fsaverage/mri/brain.nii.gz \ -out ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet_reg.nii.gz \ -omat ${SUBJECTS_DIR}/${SUBJECT}/mri/nu_bet_reg.mat \ -bins 256 -cost corratio -searchrx -90 90 -searchry -90 90 \ -searchrz -90 90 -dof 12 -interp trilinear cd $SUBJECTS_DIR tkregister2 --targ $SUBJECTS_DIR/${SUBJECT}/mri/nu_bet_reg.nii.gz \ --mov $SUBJECTS_DIR/fsaverage/mri/brain.nii.gz \ --fsl $SUBJECTS_DIR/${SUBJECT}/mri/nu_bet_reg.mat \ --reg $SUBJECTS_DIR/${SUBJECT}/mri/nu_bet_reg.dat \ --noedit -----Original Message----- From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] Sent: Monday, July 18, 2011 3:11 PM To: John Richey Cc: 'Bruce Fischl'; 'freesurfer(a)nmr.mgh.harvard.edu' Subject: Re: [Freesurfer] Problem registering labels to native space How did you create it? What is the actual "fsaverage space"? Is this an FSFAST analysis? doug John Richey wrote: > Hi Doug, the .dat file that I mention below had already been converted into > a fs matrix using tkregister2. > > -----Original Message----- > From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] > Sent: Monday, July 18, 2011 12:38 PM > To: John Richey > Cc: 'Bruce Fischl'; 'freesurfer(a)nmr.mgh.harvard.edu' > Subject: Re: [Freesurfer] Problem registering labels to native space > > You can't use a flirt matrix with mri_label2label. You can convert a > flirt matrix to the proper format with tkregister2 using the --fsl > option (this will allow you to check the reg as well). > > doug > > John Richey wrote: > >> Hi Bruce, >> >> The labels that I have fsaverage space are 3D/volume-based and not >> surface-based labels. >> >> >> -----Original Message----- >> From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] >> Sent: Monday, July 18, 2011 12:00 PM >> To: John Richey >> Cc: 'freesurfer(a)nmr.mgh.harvard.edu' >> Subject: RE: [Freesurfer] Problem registering labels to native space >> >> Hi John >> >> why are you using an affine volume registration for fsaverage? Why >> not use surface registration? >> >> Bruce >> >> >> On Mon, 18 Jul 2011, John Richey wrote: >> >> >> >>> Hi Bruce - thanks for the reply. >>> >>> Here is my command line: >>> >>> mri_label2label --srcsubject fsaverage \ --trgsubject ${SUBJECT} >>> --srclabel $SUBJECTS_DIR/group_analysis/${CONTRAST}/${LABEL} \ >>> --trglabel $SUBJECTS_DIR/${SUBJECT}/label/${CONTRAST}_${LABEL} \ >>> --regmethod volume \ --reg $SUBJECTS_DIR/$SUBJECT/mri/brain_reg.dat >>> >>> Where "brain_reg.dat" is the subject-specific matrix I created >>> through FLIRT. However, as I mentioned, disincluding this matrix >>> does not seem to change the result much. >>> >>> One other problem I encounter... Occasionally when I view the label >>> in native space, the label itself seems to be eroded on a voxel-wise >>> >>> >> basis. >> >> >>> That is, the label essentially appears as a series of non-contiguous >>> dots (which if dialated would become contiguous). >>> >>> >>> >>> >>> >>> >>> -----Original Message----- >>> From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] >>> Sent: Monday, July 18, 2011 11:20 AM >>> To: John Richey >>> Cc: 'freesurfer(a)nmr.mgh.harvard.edu' >>> Subject: Re: [Freesurfer] Problem registering labels to native space >>> >>> Hi John >>> >>> can you email us your label2label cmdline? >>> >>> cheers >>> Bruce >>> >>> >>> On Mon, 18 Jul 2011, John Richey >>> wrote: >>> >>> >>> >>>> Dear freesurfer experts, >>>> >>>> I have a series of labels corresponding to clusters of significant >>>> activation in fsaverage space. >>>> >>>> I want to map these back to native space using mri_label2label, >>>> however when I view the individual (native) labels, the >>>> registration is way off and the labels do not map into plausible space. >>>> >>>> I have tried using talairach space as an intemediary (by omitting >>>> the --reg flag in mri_label2label), as well as using fsl tools >>>> (FLIRT) to create a registration matrix but unfortunately I cannot >>>> seem to get the label into the right space to query my ces maps in >>>> > funcroi-sess. > >>>> Have others encountered this problem? And how might I create a more >>>> robust registration between fsaverage and native space to >>>> facilitate the spatial transformation of my labels? >>>> >>>> Thanks very much in advance, >>>> John >>>> _______________________________________________ >>>> 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. > >>> >>> >>> >> _______________________________________________ >> Freesurfer mailing list >> Freesurfer(a)nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> >> >> >> > > -- > Douglas N. Greve, Ph.D. > MGH-NMR Center > greve(a)nmr.mgh.harvard.edu > Phone Number: 617-724-2358 > Fax: 617-726-7422 > > Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting > FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html > > > -- Douglas N. Greve, Ph.D. MGH-NMR Center greve(a)nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html

2 1

thickness measurement workflow & tksurfer sliver
by Jaichander S 18 Jul '11

18 Jul '11

Hi freesurfers, 1. I am a new user of freesurfer. I want to measure,compare cortical thickness in corresponding regions of the cortex in a group of subjects. Is there a description of steps involved in doing this? >From what I have read till now, I know that I can create ROIs and use the instructions on http://surfer.nmr.mgh.harvard.edu/fswiki/VolumeRoiCorticalThickness to get the thickness values and then repeat the same for different regions/subjects. I would like to know if there is a workflow available for doing the same. 2. I am facing the tksurfer 'sliver' problem on a virtual CentOS machine run on Windows XP using VMWare Player for using Freesurfer. Has anyone experienced a similar problem and is there any way to correct it? Any inputs on these two issues will be very helpful. Thanks! Regards, Jaichander

1 1

Re: [Freesurfer] Problem registering labels to native space
by John Richey 18 Jul '11

18 Jul '11

Hi Doug, the .dat file that I mention below had already been converted into a fs matrix using tkregister2. -----Original Message----- From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] Sent: Monday, July 18, 2011 12:38 PM To: John Richey Cc: 'Bruce Fischl'; 'freesurfer(a)nmr.mgh.harvard.edu' Subject: Re: [Freesurfer] Problem registering labels to native space You can't use a flirt matrix with mri_label2label. You can convert a flirt matrix to the proper format with tkregister2 using the --fsl option (this will allow you to check the reg as well). doug John Richey wrote: > Hi Bruce, > > The labels that I have fsaverage space are 3D/volume-based and not > surface-based labels. > > > -----Original Message----- > From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] > Sent: Monday, July 18, 2011 12:00 PM > To: John Richey > Cc: 'freesurfer(a)nmr.mgh.harvard.edu' > Subject: RE: [Freesurfer] Problem registering labels to native space > > Hi John > > why are you using an affine volume registration for fsaverage? Why not > use surface registration? > > Bruce > > > On Mon, 18 Jul 2011, John Richey wrote: > > >> Hi Bruce - thanks for the reply. >> >> Here is my command line: >> >> mri_label2label --srcsubject fsaverage \ --trgsubject ${SUBJECT} >> --srclabel $SUBJECTS_DIR/group_analysis/${CONTRAST}/${LABEL} \ >> --trglabel $SUBJECTS_DIR/${SUBJECT}/label/${CONTRAST}_${LABEL} \ >> --regmethod volume \ --reg $SUBJECTS_DIR/$SUBJECT/mri/brain_reg.dat >> >> Where "brain_reg.dat" is the subject-specific matrix I created >> through FLIRT. However, as I mentioned, disincluding this matrix does >> not seem to change the result much. >> >> One other problem I encounter... Occasionally when I view the label >> in native space, the label itself seems to be eroded on a voxel-wise >> > basis. > >> That is, the label essentially appears as a series of non-contiguous >> dots (which if dialated would become contiguous). >> >> >> >> >> >> >> -----Original Message----- >> From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] >> Sent: Monday, July 18, 2011 11:20 AM >> To: John Richey >> Cc: 'freesurfer(a)nmr.mgh.harvard.edu' >> Subject: Re: [Freesurfer] Problem registering labels to native space >> >> Hi John >> >> can you email us your label2label cmdline? >> >> cheers >> Bruce >> >> >> On Mon, 18 Jul 2011, John Richey >> wrote: >> >> >>> Dear freesurfer experts, >>> >>> I have a series of labels corresponding to clusters of significant >>> activation in fsaverage space. >>> >>> I want to map these back to native space using mri_label2label, >>> however when I view the individual (native) labels, the registration >>> is way off and the labels do not map into plausible space. >>> >>> I have tried using talairach space as an intemediary (by omitting >>> the --reg flag in mri_label2label), as well as using fsl tools >>> (FLIRT) to create a registration matrix but unfortunately I cannot >>> seem to get the label into the right space to query my ces maps in funcroi-sess. >>> >>> Have others encountered this problem? And how might I create a more >>> robust registration between fsaverage and native space to facilitate >>> the spatial transformation of my labels? >>> >>> Thanks very much in advance, >>> John >>> _______________________________________________ >>> 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. >> >> >> >> > _______________________________________________ > Freesurfer mailing list > Freesurfer(a)nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > -- Douglas N. Greve, Ph.D. MGH-NMR Center greve(a)nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html

2 1

FreeSurfer Error during Fix Toplogy
by Kevin Liu 18 Jul '11

18 Jul '11

Hi, I'm relatively new to the FreeSurfer program and have mostly worked with processing adult brains. I am currently working on reconstructing images of young children under 5 and am consistently running into the same error for the two subjects I am running during mris_fix_topology. The error is as follows: 153 defects found, arbitrating ambiguous regions... analyzing neighboring defects... -merging segment 19 into 4 -merging segment 28 into 7 -merging segment 44 into 7 -merging segment 43 into 7 -merging segment 59 into 7 -merging segment 63 into 7 -merging segment 24 into 8 -merging segment 25 into 17 -merging segment 21 into 18 -merging segment 67 into 30 -merging segment 78 into 30 -merging segment 38 into 33 -merging segment 79 into 73 -merging segment 85 into 77 -merging segment 113 into 77 -merging segment 88 into 77 -merging segment 95 into 77 -merging segment 97 into 77 -merging segment 93 into 87 -merging segment 91 into 90 -merging segment 135 into 117 -merging segment 136 into 117 -merging segment 141 into 126 -merging segment 146 into 140 -merging segment 149 into 142 128 defects to be corrected 0 vertices coincident reading input surface /home/local/SOC/kcl388/wong_ lab/labmridata/FreeSurfer/CMH-3305/surf/lh.qsphere.nofix... reading brain volume from brain... reading wm segmentation from wm... Computing Initial Surface Statistics -face loglikelihood: -9.7664 (-4.8832) -vertex loglikelihood: -7.0295 (-3.5147) -normal dot loglikelihood: -3.5134 (-3.5134) -quad curv loglikelihood: -6.0484 (-3.0242) Total Loglikelihood : -26.3577 CORRECTING DEFECT 0 (vertices=281, convex hull=141) After retessellation of defect 0, euler #=-62 (115680,331866,216124) : difference with theory (-125) = -63 CORRECTING DEFECT 1 (vertices=100, convex hull=64) After retessellation of defect 1, euler #=-61 (115689,331920,216170) : difference with theory (-124) = -63 CORRECTING DEFECT 2 (vertices=12, convex hull=20) After retessellation of defect 2, euler #=-60 (115690,331929,216179) : difference with theory (-123) = -63 CORRECTING DEFECT 3 (vertices=395, convex hull=187) After retessellation of defect 3, euler #=-59 (115735,332152,216358) : difference with theory (-122) = -63 CORRECTING DEFECT 4 (vertices=7865, convex hull=2299) The defects never completely correct, and the program just hangs on correcting one of the defects (in this case 'DEFECT 4'). Do you have any suggestions as to where to go from here? I have checked talairach and done skull stripping edits, none of which seemed to have worked. Best, Kevin -- Kevin Liu MD Candidate 2014 Feinberg School of Medicine -- Kevin Liu MD Candidate 2014 Feinberg School of Medicine

3 3

Re: [Freesurfer] Problem registering labels to native space
by John Richey 18 Jul '11

18 Jul '11

Hi Bruce, The labels that I have fsaverage space are 3D/volume-based and not surface-based labels. -----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Monday, July 18, 2011 12:00 PM To: John Richey Cc: 'freesurfer(a)nmr.mgh.harvard.edu' Subject: RE: [Freesurfer] Problem registering labels to native space Hi John why are you using an affine volume registration for fsaverage? Why not use surface registration? Bruce On Mon, 18 Jul 2011, John Richey wrote: > Hi Bruce - thanks for the reply. > > Here is my command line: > > mri_label2label --srcsubject fsaverage \ --trgsubject ${SUBJECT} > --srclabel $SUBJECTS_DIR/group_analysis/${CONTRAST}/${LABEL} \ > --trglabel $SUBJECTS_DIR/${SUBJECT}/label/${CONTRAST}_${LABEL} \ > --regmethod volume \ --reg $SUBJECTS_DIR/$SUBJECT/mri/brain_reg.dat > > Where "brain_reg.dat" is the subject-specific matrix I created through > FLIRT. However, as I mentioned, disincluding this matrix does not seem > to change the result much. > > One other problem I encounter... Occasionally when I view the label > in native space, the label itself seems to be eroded on a voxel-wise basis. > That is, the label essentially appears as a series of non-contiguous > dots (which if dialated would become contiguous). > > > > > > > -----Original Message----- > From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] > Sent: Monday, July 18, 2011 11:20 AM > To: John Richey > Cc: 'freesurfer(a)nmr.mgh.harvard.edu' > Subject: Re: [Freesurfer] Problem registering labels to native space > > Hi John > > can you email us your label2label cmdline? > > cheers > Bruce > > > On Mon, 18 Jul 2011, John Richey > wrote: > >> Dear freesurfer experts, >> >> I have a series of labels corresponding to clusters of significant >> activation in fsaverage space. >> >> I want to map these back to native space using mri_label2label, >> however when I view the individual (native) labels, the registration >> is way off and the labels do not map into plausible space. >> >> I have tried using talairach space as an intemediary (by omitting the >> --reg flag in mri_label2label), as well as using fsl tools (FLIRT) to >> create a registration matrix but unfortunately I cannot seem to get >> the label into the right space to query my ces maps in funcroi-sess. >> >> Have others encountered this problem? And how might I create a more >> robust registration between fsaverage and native space to facilitate >> the spatial transformation of my labels? >> >> Thanks very much in advance, >> John >> _______________________________________________ >> 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. > > >

2 1

Re: [Freesurfer] Problem registering labels to native space
by John Richey 18 Jul '11

18 Jul '11

Hi Bruce - thanks for the reply. Here is my command line: mri_label2label --srcsubject fsaverage \ --trgsubject ${SUBJECT} --srclabel $SUBJECTS_DIR/group_analysis/${CONTRAST}/${LABEL} \ --trglabel $SUBJECTS_DIR/${SUBJECT}/label/${CONTRAST}_${LABEL} \ --regmethod volume \ --reg $SUBJECTS_DIR/$SUBJECT/mri/brain_reg.dat Where "brain_reg.dat" is the subject-specific matrix I created through FLIRT. However, as I mentioned, disincluding this matrix does not seem to change the result much. One other problem I encounter... Occasionally when I view the label in native space, the label itself seems to be eroded on a voxel-wise basis. That is, the label essentially appears as a series of non-contiguous dots (which if dialated would become contiguous). -----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Monday, July 18, 2011 11:20 AM To: John Richey Cc: 'freesurfer(a)nmr.mgh.harvard.edu' Subject: Re: [Freesurfer] Problem registering labels to native space Hi John can you email us your label2label cmdline? cheers Bruce On Mon, 18 Jul 2011, John Richey wrote: > Dear freesurfer experts, > > I have a series of labels corresponding to clusters of significant > activation in fsaverage space. > > I want to map these back to native space using mri_label2label, > however when I view the individual (native) labels, the registration > is way off and the labels do not map into plausible space. > > I have tried using talairach space as an intemediary (by omitting the > --reg flag in mri_label2label), as well as using fsl tools (FLIRT) to > create a registration matrix but unfortunately I cannot seem to get > the label into the right space to query my ces maps in funcroi-sess. > > Have others encountered this problem? And how might I create a more > robust registration between fsaverage and native space to facilitate > the spatial transformation of my labels? > > Thanks very much in advance, > John > _______________________________________________ > 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.

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