Freesurfer

freesurfer@nmr.mgh.harvard.edu

9 participants
30092 discussions

Re: [Freesurfer] Fw: PETsurfer mri_coreg problem
by Antonin Skoch 26 Feb '17

26 Feb '17

Dear Paul, LTA means transformation matrix stored in .lta format. This matrix can encode transformation between various coordinate spaces. What coordinate systems transform this .lta from and to is specified by string in the .lta file. Typically this type could be vox2RAS, RAS2vox, vox2vox or RAS2RAS. Most typically the type is vox2vox. The explanation of coordinate systems is specified for example here: http://freesurfer.net/fswiki/CoordinateSystems http://www.grahamwideman.com/gw/brain/fs/coords/fscoords.htm The output of freeview is only info, not error, indicating that your .lta is not of type vox2vox. I personally still use tkregister2 for checking of registration, since the freeview does not contain full functionality for checking transformations (see the warning). Therefore I have no experience with tkregisterfv and consider tkregister2 still as a "gold standard". However, if you use tkregisterfv for checkning transformation between volumes and the output is only "info", I would guess that everything is OK and you do not need to be bothered. Btw, does the alignment of the images seem reasonable? How does it look in tkregister2? Regards, Antonin

3 2

PETsurfer mri_coreg problem
by miracooloz＠gmail.com 26 Feb '17

26 Feb '17

2 2

CORTICAL THICKNESS
by Flavia Arduini 26 Feb '17

26 Feb '17

Hi everyone, I segmented with Mantis, a toolbox of spm, MRI T2-weighted images of infants . Now I have to compute cortical thickness. How can I do it with Freesurfer? I know there's the recon-all function that segments and does this but is based on T1-weighted images. There is a way to be able to adapt? Thank you and I hope someone can help me. Flavia Arduini

1 0

Re: [Freesurfer] Mindboggle official software release and publication!
by Antonin Skoch 26 Feb '17

26 Feb '17

Dear experts, I was wondering what is your opinion on the fact, that, according to the mindboggle paper http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.10053… it is beneficial to use ANTS volume-based gray matter segmentation to refine gray matter segmentation in FreeSurfer. I am not expert in ANTS but looking at the ANTS code it seems to me that the ANTS routine does not do anything more than the volume-based nonlinear registration to the template and intensity and probabilistic atlas-based voxel labeling. Therefore, the method is in principle identical to GCA volume-based labeling (mri_ca_register and mri_ca_label, part of recon-all) in FreeSurfer which also produces cortical gray matter mask, far not so precise as the segmentation based on pial and white surface estimation. I think that FreeSurfer pipeline for estimation of white and pial surfaces is much more sophisticated and precise than any method using volume-based registration and volume-based labeling by intensity and anatomical priors. The FreeSurfer gray matter underestimation showed in figure 2 of the paper http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.10053… seems to me like either partial-volume effect or skull-strip error (which cuts out part of gray matter) which can be corrected by proper inspection and manual correction of brainmask.mgz. In contrast, the errors in presented ANTS gray matter segmentation in figure 2 are much more severe, spanning not only the blue encircled region. I would very appreciate your expert opinion on this. Regards, Antonin Skoch Announcing the official release of Mindboggle (http://mindboggle.info) open source software <http://mindboggle.info/software.html> and data <https://osf.io/ydyxu/> for analyzing the shapes of brain structures from human MRI data (processed through FreeSurfer and optionally through ANTs). The release coincides with a publication in PLoS Computational Biology that documents and evaluates the software:Klein A, Ghosh SS, Bao FS, Giard J, Hame Y, Stavsky E, Lee N, Rossa B, Reuter M, Neto EC, Keshavan A. (2017) Mindboggling morphometry of human brains. PLoS Computational Biology 13(3): e1005350. doi:10.1371/journal.pcbi.1005350 <http://doi.org/10.1371/journal.pcbi.1005350> Cheers, @rno

1 0

Disastrous Soft Error
by Rockers, Elijah D. 25 Feb '17

25 Feb '17

Hello all, I recon-all'd ~40 different subjects. I have been going through all the images to make sure they were processed correctly. Most required no no editing and recon-all did a very good job. But for one scan in particular, it looks like Freesurfer thinks the top of the spinal cord is the brain. Several vertebrae are very nicely segmented... :P Just wondering if you have any suggestions for reprocessing this subject? The error is reproducible with default recon-all settings. Subject has some pathology in the white matter which might be causing the problem? Screenshot of T1.mgz + lh.pial at the following link: http://imgur.com/a/uwyaq Thanks, Eli Houston Methodist. Leading Medicine. U.S. News & World Report has recognized Houston Methodist as one of the nation’s top 20 hospitals, placing it for the second time on the magazine’s prestigious Honor Roll. It is also designated as a Magnet hospital for excellence in nursing. Visit us at houstonmethodist.org. Follow us at twitter.com/MethodistHosp and www.facebook.com/HoustonMethodist. ***CONFIDENTIALITY NOTICE*** This e-mail is the property of Houston Methodist Hospital and/or its relevant affiliates and may contain restricted and privileged material for the sole use of the intended recipient(s). Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender and delete all copies of the message. Thank you.

2 3

Re: [Freesurfer] lh.white.H , Failed to create
by Antonin Skoch 25 Feb '17

25 Feb '17

Dear Vasudev, on what filesystem does your working directory reside? According to the path to your data it seems that you use some external drive, probably with non-unix-like filesystem. The error most probably means, that your filesystem does not support symbolic links. Regards, Antonin Dear all , I have been incurring lot of problems while running recon-all on ubuntu 16.04 , recon-all throws following error : " #@# Curv .H and .K lh Sa 25. Feb 04:09:16 CET 2017 /media/vasudev/vkirsch4/Project_BLVP/HCP_Proj/KON2/T1w/KON2/surf mris_curvature -w lh.white.preaparc total integrated curvature = 8.913*4pi (112.005) --> -8 handles ICI = 172.8, FI = 1966.7, variation=30317.492 writing Gaussian curvature to ./lh.white.preaparc.K...done. writing mean curvature to ./lh.white.preaparc.H...done. rm -f lh.white.H ln -s lh.white.preaparc.H lh.white.H ln: failed to create symbolic link 'lh.white.H': Function not implemented " Could you please let me know why this error is occurring . Thanks Vasudev

1 0

lh.white.H , Failed to create
by Dev vasu 25 Feb '17

25 Feb '17

Dear all , I have been incurring lot of problems while running recon-all on ubuntu 16.04 , recon-all throws following error : " #@# Curv .H and .K lh Sa 25. Feb 04:09:16 CET 2017 /media/vasudev/vkirsch4/Project_BLVP/HCP_Proj/KON2/T1w/KON2/surf mris_curvature -w lh.white.preaparc total integrated curvature = 8.913*4pi (112.005) --> -8 handles ICI = 172.8, FI = 1966.7, variation=30317.492 writing Gaussian curvature to ./lh.white.preaparc.K...done. writing mean curvature to ./lh.white.preaparc.H...done. rm -f lh.white.H ln -s lh.white.preaparc.H lh.white.H ln: failed to create symbolic link 'lh.white.H': Function not implemented " Could you please let me know why this error is occurring . Thanks Vasudev

1 0

recon-all error (6.0)
by Trisanna Sprung-Much 24 Feb '17

24 Feb '17

Hi All I am trying to run a subject in 6.0 (it ran perfectly well in 5.3) and it stops when trying to transform to the "conformed" space. I get the following error after about a minute. I have not seen this error before. mri_convert ./tmp.mri_nu_correct.mni.2860/nu1.mnc orig_nu.mgz --like orig.mgz --conform mri_convert.bin ./tmp.mri_nu_correct.mni.2860/nu1.mnc orig_nu.mgz --like orig.mgz --conform ncopen: filename "./tmp.mri_nu_correct.mni.2860/nu1.mnc": Not a netCDF file miopen: MINC package entry point Error opening ./tmp.mri_nu_correct.mni.2860/nu1.mnc mincRead(): error reading volume from file ./tmp.mri_nu_correct.mni.2860/nu1.mnc $Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $ reading from ./tmp.mri_nu_correct.mni.2860/nu1.mnc... Linux kaplan 3.13.0-105-generic #152-Ubuntu SMP Fri Dec 2 15:37:11 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux recon-all -s icbm-131 exited with ERRORS at Thu Feb 23 11:40:53 EST 2017 To report a problem, see http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting Any ideas? Best Trisanna -- Ph.D. Candidate McGill University Integrated Program in Neuroscience Psychology

2 2

Mindboggle official software release and publication!
by Arno Klein 24 Feb '17

24 Feb '17

Announcing the official release of Mindboggle (http://mindboggle.info) open source software <http://mindboggle.info/software.html> and data <https://osf.io/ydyxu/> for analyzing the shapes of brain structures from human MRI data (processed through FreeSurfer and optionally through ANTs). The release coincides with a publication in PLoS Computational Biology that documents and evaluates the software: Klein A, Ghosh SS, Bao FS, Giard J, Hame Y, Stavsky E, Lee N, Rossa B, Reuter M, Neto EC, Keshavan A. (2017) Mindboggling morphometry of human brains. PLoS Computational Biology 13(3): e1005350. doi:10.1371/journal.pcbi.1005350 <http://doi.org/10.1371/journal.pcbi.1005350> Cheers, @rno

4 4

mri_segstats
by John Anderson 24 Feb '17

24 Feb '17

Dear Freesurfer experts, I have used the following commandline to generate the final statistics in my analysis: mri_segstats --seg aseg.mgz --ctab-default --i suvr.nii --mask mask.nii --sum stats.dat The final output of this command is (attached). Depedning on the anatomical location of the mask, I expect that I will get two structures in the stats file. What is the "unknown" label means? is it CSF? Do I need to include it in my next statistcal analses? I highly appreciate your input! Best, John Sent with [ProtonMail](https://protonmail.com) Secure Email.

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