Freesurfer

freesurfer@nmr.mgh.harvard.edu

16 participants
30084 discussions

"wm" boundaries and stats
by Krieger, Donald N. 03 Mar '16

03 Mar '16

Freesurfer provides 34 ctx label volumes on each side of the brain with adjacent wm labels. The boundary between each ctx volume and the co-named adjacent wm volume is a third "white" label. I would like to extract a few pieces of information from the analogous boundary between each wm volume and those adjacent volumes to it which are not the co-named ctx volume. For instance many wm volumes share a boundary with the unsegmented white matter. Some share a boundary with sub-cortical structures. Here are some of the information I would like to extract. What do you think? What are the xyz coordinates of the complete boundary surrounding each wm volume and what is the name of the adjacent volume for each point on the boundary? I understand that for the boundary between the ctx volume and the adjacent wm volume, the white label provides this information. What is the percentage and area of the complete boundary surrounding a wm volume which is adjacent to the co-named ctx volume.? What are the names of the other volumes with which each wm volume shares a boundary and what is the percentage and area of the boundary shared with each? For each point within a wm volume, are the nearest points on the wm boundary adjacent to the co-named ctx volume or are some adjacent to unsegmented white matter or sub-cortical structures? More generally, is a particular few mm voxel surrounded by gray matter and so is likely primarily composed primarily of relatively short range fibers or is it adjacent to the deep white matter (unsegmented) and likely composed of a mix of relatively short range fibers and long range "en passage" fibers. It is this last question which most closely gets at the ideas I wish to explore. We are able to measure electric current in the white matter with mm resolution from the MEG. The mix of local and "en passage" fibers should effect the direction of the electric currents in the white matter and I can measure that. The distance scale is at the edge of that provided by DTI-derived tractography so we may be able to meld these measurements with those. Your thoughts would be welcome on what is already there in freesurfer that will help with this and what you think might be a good direction to proceed to develop what we need from what is there. Thanks - Don

2 2

v5.3 fsaverage (accidentally) in v5.1 directory
by Adam Martersteck 03 Mar '16

03 Mar '16

Hi Freesurfer team, *Short question: *Will (accidentally) using the fsaverage v5.3 subject in our v5.1 directory influence our reconstructions? *Long question:* Our server has 2 different freesurfer environments. A recon was accidentally run on a single v5.1 subject in the v5.1 directory with freesurfer version 5.3. During the “balabels” stage of autorecon3, because FS 5.3 couldn’t find the perirhinal label – it deleted our 5.1 symlink and replaced it with the correct 5.3 symlink. We realized it had happened and correct the subject, but didn’t catch that the symlink had been changed. We have run multiple (hundreds) of brains through v5.1 since then – with the v5.3 fsaverage in the v5.1 directory. A cursory glance over the release notes from 5.1 to 5.3 makes me think nothing has fundamentally changed – only new files have been added to fsaverage 5.3 (e.g. cortical parcellation networks). No new GCA atlas or anything that would change how FS 5.1 reconstructs the brain? Thanks for any thoughts! Adam

2 1

conflicting Monte Carlo simulation results...
by Karl Liu 03 Mar '16

03 Mar '16

Dear Freesurfer experts: I ran into a confusing problem when applying Monte Carlo simulation to control for multiple comparisons. I am using the buckner data from tutorial dataset. I first tried the simulation based on Freesurfer;s 2012 manual: (http://www.freesurfer.net/pub/docs/tarballs_for_courses/FreeSurferTutorial-…) mri_glmfit-sim \ --glmdir lh.gender_age.glmdir \ --sim mc-z 5 4 mc-z.negative \ --sim-sign neg --cwpvalthresh 0.05\ --overwrite This produced a "mc-z.negative.sig.cluster.summary". In this text, no clusters survived correction. Later on, I applied on a slightly different script, based on Freesurfer tutorial (http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis#Clusterwi…) mri_glmfit-sim \ --glmdir lh.gender_age.glmdir \ --cache 4 neg \ --cwp 0.05\ --2spaces It produced a "cache.th40.neg.sig.cluster.summary" this time. But according to this result, 22 clusters survived. Now I am quite confused, because both corrections have the same voxel-wise (p=0.001) and cluster-wise (p=0.05) threshold. Where the drastic discrepancies come from and which result is more reliable? Many thinks in advance. Best, Karl

2 1

Extract LGI at MaxVertex and extract effect-sizes in table format
by Bittner, Nora 03 Mar '16

03 Mar '16

Dear FreeSurfer Experts, Sorry for reposting this massage again, but I still have two questions, which you can hopefully answer to: 1) I calculated several analyses with LGI as my dependent variable, resulting in several significant clusters. Now, I would like to extract the individual LGI value at the maximum vertex of each significant cluster to plot it against my experimental variables (instead of working with the mean LGI values of the clusters). What I tried so far is to use mri_segstats to extract the LGI at the maximum vertex: mri_segstats --i $SUBJECTS_DIR/qdec/glm/y.mgh --crs vertexno 84367 0 0 --avgwf vertexno84367.dat where y.mgh is my input file from the glm and vertexno 84367 is the maximum vertex of the cluster as reported in the cluster.summary-file. I found the command in the FS helpline documentation (https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg42655.html) but regrettably<http://dict.leo.org/ende/index_en.html#/search=regrettably&searchLoc=0&resu…> I get the following error message: "ERROR: Option --crs unkown" In the fswiki I also couldn´t find the "mri_segstats - - crs option" in the user description. Is there any other option or argument I can use in mri_segstats (or in any other command) to extract the LGI value per person at the given vertex? 2) I would like to calculate effect sizes of my results. Therefore I used fscalc: fscalc $SUBJECTS_DIR/qdec/glm/contrast/gamma.mgh div $SUBJECTS_DIR/qdec/glm/rstd.mgh -o $SUBJECTS_DIR/qdec/ glm/contrast/effectsize.mgh This worked fine and resulted in a surface overlay displaying the effect sizes. Now, I would like to extract these effect sizes in 'table' format. I tried to use mri-segstats: mri_segstats* --i effectsize.mgh --annot fsaverage lh aparc --avgwf effectsize.dat I am not sure, if this gives me the correct results. Is this the right command? I would really appreciate<http://dict.leo.org/ende/index_en.html#/search=appreciate&searchLoc=0&resul…> your help. Thank you in advance, Nora ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Forschungszentrum Juelich GmbH 52425 Juelich Sitz der Gesellschaft: Juelich Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498 Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender), Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt, Prof. Dr. Sebastian M. Schmidt ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------

2 1

Reading values from overlays with PETcoreg
by Benjamin Spurny 03 Mar '16

03 Mar '16

Dear Freesurfer experts! I am currently working on PET analysis using FS I coregistered my PET with the processed MR using bbregister, transfered it to a surface using mri_vol2surf and now createt an overlay in freeview with the lh.inflated and used the labels from the lh.aparc.a2009s.annot file. In freeview i get the corresponding BP value for each vertex now but is there a way to get a list of vertices with the corresponding BP value and the corresponding ROI this vertex belongs to? Or is there a better to do this analyis? Many thanks in advance! Benjamin

3 14

[FreeSurfer] Repeated Measures ANOVA in FSFAST
by Afzal, Afsana 03 Mar '16

03 Mar '16

Hi, I'm analyzing a decision making task with a 3x3 condition matrix: pairings of low, medium, high levels of risk with low, medium, high levels of reward. I've set up the following FSGD file to do a repeated measures/within-subject ANOVA: GroupDescriptorFile 1 Class hc001 Class hc002 Class hc003 Class hc004 Class hc005 Variables lRsk_lRwd lRsk_mRwd lRsk_hRwd mRsk_lRwd mRsk_mRwd mRsk_hRwd hRsk_lRwd hRsk_mRwd hRsk_hRwd Input hc001 hc001 1 1 1 1 1 1 1 1 1 Input hc002 hc002 1 1 1 1 1 1 1 1 1 Input hc003 hc003 1 1 1 1 1 1 1 1 1 Input hc004 hc004 1 1 1 1 1 1 1 1 1 Input hc005 hc005 1 1 1 1 1 1 1 1 1 I've also created the following contrast matrices: Con1: Does difference between highRisk and LowRisk differ from 0? 0 0 0 0 0 -1 -1 -1 0 0 0 1 1 1 Con2: Does difference between highRisk and MedRisk differ from 0? 0 0 0 0 0 0 0 0 -1 -1 -1 1 1 1 Is this an appropriate way of setting up the ANOVA? Thanks for your help, Afsana ______________________________ Afsana Afzal Clinical Research Coordinator Massachusetts General Hospital Division of Neurotherapeutics Department of Psychiatry: Neurosciences 149 13th St, Room 2612 Charlestown, MA 02129 Phone: 617-643-5129 Fax: 617-726-4078

2 5

Topological White Matter Correction
by Kiersten Snyder 02 Mar '16

02 Mar '16

Hello FreeSurfer Experts, What image processing technique is behind the Topological White Matter Correction to correct topological defects? Is it true that FreeSurfer take into account your edits as a "suggestion" and incorporates into some type of algorithm? I would like to have a more thorough understanding of the image processing that this function utilizes. Thanks for the help! Kiersten

2 1

hippocampus subfields
by Kanel, Prabesh 02 Mar '16

02 Mar '16

Freesurfer List, I ran recon-all to segment hippocampus subfield using the development version of freesurfer.V6.Beta. To get the better segmentation, I have used both T1 and T2 images. But for some of the scans, the process stops at a point and doesn't seem to progress further. That happens in only a few cases and only when the data are run using the flag -hippocampal-subfields-T1T2. I have attached a log file of one of the case in this email. Another query that I have is to get T2 segmented subfields images (one with intensity values rather than labels). Is there an easier way to get T2 subfields segmented images using the segmentation volumes and T2.FSspace.mgz. Sorry, if this question seems too trivial. Regards, Prabesh

2 4

Unknown option DKTatlas40
by marmaduke woodman 02 Mar '16

02 Mar '16

hi I'm using the dev version of FreeSurfer on 64-bit Fedora 23. Recon-all runs fine but at the end I see $ recon-all -s tvb -aparc2aseg ... mri_aparc2aseg --s tvb --volmask --aseg aseg.presurf.hypos --DKTatlas40 ERROR: Option --DKTatlas40 unknown Running that command $ mri_aparc2aseg --s tvb --volmask --aseg aseg.presurf.hypos without the --DKTatlas40 appears to run correctly. Is this just an issue with the dev version or have I done something silly? thanks, Marmaduke ps bugr output FREESURFER_HOME: /opt/freesurfer Build stamp: freesurfer-Linux-centos6_x86_64-dev-20160229 RedHat release: Fedora release 23 (Twenty Three) Kernel info: Linux 4.3.3-300.fc23.x86_64 x86_64

3 2

Re: [Freesurfer] running two versions of fs in the same
by Fengji Geng 01 Mar '16

01 Mar '16

Hi Bruce, The command that we ran is recon-all. Below is the error msg: mri_label2label --srcsubject fsaverage --srclabel /export/data/hippocampus/HMN/data/fsaverage/label/lh.BA2.label --trgsubject HMN122 --trglabel ./lh.BA2.label --hemi lh --regmethod surface No such file or directory mri_label2label: could not open label file /export/data/hippocampus/HMN/data/fsaverage/label/lh.BA2.label srclabel = /export/data/hippocampus/HMN/data/fsaverage/label/lh.BA2.label srcsubject = fsaverage trgsubject = HMN122 trglabel = ./lh.BA2.label regmethod = surface srchemi = lh trghemi = lh trgsurface = white srcsurfreg = sphere.reg trgsurfreg = sphere.reg usehash = 1 Use ProjAbs = 0, 0 Use ProjFrac = 0, 0 DoPaint 0 SUBJECTS_DIR /export/data/hippocampus/HMN/data FREESURFER_HOME /cell_root/software/freesurfer/current/sys Loading source label. Invalid argument ERROR reading /export/data/hippocampus/HMN/data/fsaverage/label/lh.BA2.label Linux hippocampus.umd.edu 2.6.32-573.12.1.el6.x86_64 #1 SMP Mon Nov 23 12:55:32 EST 2015 x86_64 x86_64 x86_64 GNU/Linux recon-all -s HMN122 exited with ERRORS at Tue Feb 23 16:40:01 EST 2016 ---------------------------------------------------------------------------------- Hi Fengji you need to give us more details about the problems before we can help you. What commands failed? What was the screen output? Note that there is still no such thing as FS 6.0. We are hoping to have it in the relatively near term, but it doesn't exist yet. cheers Bruce On Tue, 1 Mar 2016, Fengji Geng wrote: > Dear fs developers,? > > we are attempting to use the stable version of fs 5.1 for processing while > also using the most current version for hippocampal segmentation on the same > set of subjects (freesurfer 6.0 dev version on 28-Jan-2016). However, when > the develop version of fs 6 was installed, we began having an issue with the > fsaverage folder linking to the newest version, when we need it to link to > version 5.1 during processing. The label files that we need for processing > in version 5.1 are not accessible in the new version?s fsaverage > folder.?However, before we updated to fs 6.0 dev version on 28-Jan-2016, we > never had this issue. So, we are wondering if the newest develop version of > fs 6.0 does not allow?two versions of fs to be used in the same data.? > > Looking forward to hearing from you and thanks in advance. > > Best, > > ----- Fengji --- Fengji

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