Freesurfer August 2021

freesurfer@nmr.mgh.harvard.edu

73 participants
101 discussions

SAMSEG esitmation error
by achille.teillac＠chu-lyon.fr 06 Sep '21

06 Sep '21

External Email - Use Caution Dear experts, We are working with a large MRI dataset (FLAIR, T1) and we are performing SAMSEG pipeline to estimate lesion volumes. We computed the registration between the two input images and the resampling as advised in the SAMSEG page (mri_coreg and mri_vol2vol) which worked perfectly. However, when we do the visual chek on the segmentation after running SAMSEG, it is completely wrong for some subjects. By looking at corresponding mode01_bias_corrected.mgz and mode02_bias_corrected.mgz images, they are too cropped (centered on the cerebellum) compared to the original ones (3D_T1 .nii.gz and 3D_FLAIR.nii.gz). We also noticed a < RuntimeWarning: invalid value encountered in true_divide > during the burn-in phase (in the GMM.py code). Do you have any clue of what is happening with this cropping (can it be linked to the bias field estimation ?) and how we can correct it ? Thanks for your help and I can provide some screenshots of the cropped images to help visualize the issue. Best, Achille Teillac

2 1

fs_install permission denied
by Junyeon Won 02 Sep '21

02 Sep '21

External Email - Use Caution Dear Freesurfer experts, I was tiring to install fs_install using the below command. cd $FREESURFER_HOME/bin && curl https://secure-web.cisco.com/186mRL7xDETIpN5cZS-wivuTmyhZn0PmwqT8bMpnWm3D4D… -o fs_install_mcr && chmod +x fs_install_mcr It showed an error message saying permission denied so I added sudo as suggested here. https://secure-web.cisco.com/1d_mS4e4I0j64PnpWDAM0mBO-6wAmGqcg5riZxWWy6VAP8… However, even after adding sudo in front, it shows the same error message. % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 2010 100 2010 0 0 17104 0 --:--:-- --:--:-- --:--:-- 17179 chmod: Unable to change file mode on fs_install_mcr: Operation not permitted Could you please let me know how to resolve this? Best, JW ᐧ

2 2

recon-all Longitudinal processing (baseline on 1.5T; followup on 3T)
by Jayachandra Raghava 02 Sep '21

02 Sep '21

External Email - Use Caution Hello Experts, I am planning to run the longitudinal processing pipeline (Freesurfer version 7.1.1). But I have a few subjects who were scanned on different scanner field strengths at baseline and followup. For example : Baseline was scanned on a 1.5T Siemens Scanner Followup was scanned on a 3.0T Siemes Trio Scanner I am planning to 1) run the cross-sectional processing for baseline without the 3T flag 2) run the cross-sectional processing for followup with the 3T flag 3) then generate the unbiased template and longitudinally process all the time points without any special flags. Is this the best way to process such data? Please let me know if there are any additional flags in the longitudinal processing pipeline to accommodate Scanner/Field strength changes. Best wishes Jay

3 2

Shiny new 7.2!
by Andy Worth 01 Sep '21

01 Sep '21

External Email - Use Caution Is it true that all that is needed for a new official FS docker container to come into existence is to merge pull request #866 <https://secure-web.cisco.com/10DeE1N6s0_SCCbPo6uqcazn6bfeF4GN2L9n26obY0mrz4…>? That would be so very nice. We like the official release! 4.1K pulls of previous versions on dockerhub can't be wrong. Kind regards, Andy Worth Senior Scientific Solutions Engineer he, him, his <http://secure-web.cisco.com/1q0M-qEiBJlwu49m9WEWx5rGqctEdeZiwaiHNtscHhu8sb1…>

2 1

TOPO_DEF
by halil ibrahim AKÇAY 31 Aug '21

31 Aug '21

External Email - Use Caution Hi My name is Halil Ibrahim akçay. I am a neurologist. I have been used FreeSurfer for a while. And I cannot understand some part of troubleshooting process. I am sending a screenshot of my question. How can I understand whether gray zone as marked blue circle is a problem or not in this picture? And Should I fill this zone to fix the problem, if this is a problem? Thanx for your helping. I am looking forward to your answer. Best regards

2 1

ERROR: mri_ca_register with non-zero status 11
by 朱子家 31 Aug '21

31 Aug '21

External Email - Use Caution Hello FreeSurfer Developers, when I run the recon-all command: $ export SUBJECTS_DIR=`pwd` $ cd /lustre/home/acct/zjzhu/project/zzu_MRI/T1/test $ ls *nii | parallel --jobs 12 recon-all -s {.} -i {} -all -qcache I get the following error ERROR: mri_ca_register with non-zero status 11 recon-all -s co20160131_165733 exited with ERRORS at Mon Aug 30 13:57:15 CST 2021 I've searched the list and no similar errors have been reported. Does anyone have any thoughts on how to trouble-shoot this one? Also, Ive attached the recon-all.log in case it's of any use. 1) FreeSurfer version: freesurfer-linux-centos7_x86_64-7.1.1-20200723-8b40551 2) Platform: CentOS Linux release 7.7.1908 (Core) 3) Linux login1.pi.sjtu.edu.cn 3.10.0-1062.el7.x86_64 #1 SMP Wed Aug 7 18:08:02 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux 4) recon-all.log: see attached

2 1

Re: [Freesurfer] recon-all exited with ERRORS
by guang 31 Aug '21

31 Aug '21

External Email - Use Caution At 2021-08-05 08:44:11, "guang" <zhaoguang721(a)163.com> wrote: Hello FreeSurfer Developers, I'm trying to analyze aseg.mgz files for each subjects though i have segmented hippocampal subfields and nuclei of the amygdala. Can I make the voxel-by voxel group analysis with the aseg.mgz or the aparc+aseg.mgz files? Thank you for any reply. guang

2 1

Freesurfer Matlab LME tool: problem with function lme_mass_RgGrow
by Bonati, Sara 31 Aug '21

31 Aug '21

External Email - Use Caution Dear FreeSurfer Developers, I'm attempting to analyze thickness changes in a longitudinal design using a mass-univariate approach, as described on the Linear Mixed Effects tutorial at https://secure-web.cisco.com/1HwLf5wG3aIAhV84T5KTL4VUiVpoa1DISeYUypdAqYlExO…. Before starting the tutorial I followed the preprocessing steps described in section 2 at https://secure-web.cisco.com/1DrlLxryQICRHsEOZazZuvsDg0-s9654AX1e3qQgZGXx_G… I now have a dataset folder that contains 1) for each subject-session pair to include in the analysis a .long folder 2) a subject specific base template folder that used all available sessions from that subject. All recon-all.log files, both in the base template folder and the subject-session specific long folders, seem to have finished without error. After that, to prepare files for LME, I created a longitudinal qdec file and used it to prepare thickness.mgh files as described in the beginning of the Linear Mixed Effects tutorial for the mass-univariate approach. When advancing in the analysis the function lme_mass_RgGrow gets stuck: in the terminal the output looks like this ``` Starting region growing algorithm Computing seeds ... ``` and stays like this without advancing. I tried to change the thickness file used (lh.thickness.mgh VS lh.thickness_sm5.mgh VS lh.thickness_sm10.mgh etc..) but unfortunately the error persists. Looking in the mailing list I found an email with a problem that is similar to mine: https://secure-web.cisco.com/166fF5937wdn5yDLqNvdeV6NeSe3cGRWXBnp12bBcQPqvc… . Upon inspection of my lh.thickness.mgh file I also noticed that some columns are all zeros: do you know what could cause such columns of zeros to appear in the thickness files and how the problem with the function could be solved to continue in the analysis? Thank you very much for your help, Kind regards Sara Bonati 1) FreeSurfer version: freesurfer-darwin-macOS-7.2.0-20210713-aa8f76b 2) Platform: macOS Catalina version 10.15.7

2 1

mri is cut off
by Ricardo Adrogue 31 Aug '21

31 Aug '21

External Email - Use Caution Hi y,all, I know that freesurfer is not made to work with MRI scans that are cut off. However, this is the only imaging we have, and we would love to have freesurfer data to analyze from this patient. The posterior third of the scan is missing, so it will not be an easy fix. If anyone has any suggestions on things to try, please let me know. Thanks! Ricardo Adrogue

1 0

Re: [Freesurfer] SAMSEG esitmation error
by Stefano Cerri 31 Aug '21

31 Aug '21

External Email - Use Caution Hi Achille, It sounds like the initial affine registration of the template in SAMSEG is failing for some of your subjects. If that is the case, you will see that "template_coregistered.nii" is probably not aligned with your original data. As the affine registration has been improved in the latest version (FS 7.2), can you try to run a couple of failed segmentations on this new version and see if it solves the issue? If that doesn't work, it would be good if you can send us a couple of representative subjects so that we can see where SAMSEG fails. Best, Stefano Dear experts, We are working with a large MRI dataset (FLAIR, T1) and we are performing SAMSEG pipeline to estimate lesion volumes. We computed the registration between the two input images and the resampling as advised in the SAMSEG page (mri_coreg and mri_vol2vol) which worked perfectly. However, when we do the visual chek on the segmentation after running SAMSEG, it is completely wrong for some subjects. By looking at corresponding mode01_bias_corrected.mgz and mode02_bias_corrected.mgz images, they are too cropped (centered on the cerebellum) compared to the original ones (3D_T1 .nii.gz and 3D_FLAIR.nii.gz). We also noticed a < RuntimeWarning: invalid value encountered in true_divide > during the burn-in phase (in the GMM.py code). Do you have any clue of what is happening with this cropping (can it be linked to the bias field estimation ?) and how we can correct it ? Thanks for your help and I can provide some screenshots of the cropped images to help visualize the issue. Best, Achille Teillac

1 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer August 2021