Freesurfer

freesurfer@nmr.mgh.harvard.edu

16 participants
30084 discussions

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

See masked area in freeview
by Haley Park 31 Aug '21

31 Aug '21

External Email - Use Caution Hello, I did a correlation analysis and see the significant areas in freeview when I run freeview -f $SUBJECTS_DIR/fsaverage/surf/lh.inflated:overlay=cache.th20.abs.sig.cluster.mgh However, I would like to see whether the significant areas are a part of the Yeo 2011 7 Networks. https://secure-web.cisco.com/1rw_v7ySWe7KBTv8MIybez7KI7Zn2aGlwvG8tPfwhThugk… Is there a way to overlay the Yeo network area in Freeview? Or some other way to see if the significant areas are a part of the network and visualize it? Thank you, Haley

2 3

Segmentation fault
by Nicole Espinosa Zarlenga 30 Aug '21

30 Aug '21

External Email - Use Caution Hi, I am trying to use Freeview in my MacBook Pro. However, every time I tried to open it from terminal I get this error: /Applications/freesurfer/7.2.0/bin/freeview: line 2: 5984 Segmentation fault: 11 $FREESURFER_HOME/Freeview.app/Contents/MacOS/freeview "$@" I Freeview closes automatically. Do you know how I can fix this error? Thanks! Nicole Espinosa Research Assistant | School of Psychology | Faculty of Science THE UNIVERSITY OF SYDNEY Healthy Brain Ageing Program Brain and Mind Centre & Charles Perkins Centre E: nesp6016(a)uni.sydney.edu.au<mailto:nesp6016@uni.sydney.edu.au>

1 0

Postdoctoral Fellow of Neuroimaging at Emory University
by Wu, Junjie 30 Aug '21

30 Aug '21

External Email - Use Caution The Department of Radiology and Imaging Sciences of Emory University is seeking two postdoctoral fellow to work on the development and/or application of advanced MRI techniques in Alzheimer’s Disease. The postdoctoral fellows will join a team of investigators in the large-scale NIH-funded Emory Healthy Brain Study, which is developing early predictive biomarkers for Alzheimer’s Disease with a target enrollment of 3000 subjects ( https://secure-web.cisco.com/1LADw4apndXUutJQ9dWviN6_CQ02Qv0_Bgj5FKHtemOE55…) The position provides an excellent opportunity to work in an interdisciplinary environment with close interactions with MR physicists, radiologists, neurologist, neuroscientists, and computer scientists. The Center for Systems Imaging at Emory University operates three research-dedicated 3T MRI Siemens Prisma scanners one of which is exclusively used for the Emory Healthy Brain Study, one human PET/MR scanner, and a powerful computer-cluster. A Siemens 7T Terra human scanner will be installed in early 2022 for research use. The position holder is expected to assist with MR data acquisition, performing image analysis and writing manuscripts. Several advanced MR sequences are acquired in the Emory Healthy Brain Study, including resting state fMRI, diffusion MR, arterial spin labeling, quantitative susceptibility mapping, MR spectroscopy among others. The exact area of focus for the Postdoctoral Fellow will depend on their existing experience and programmatic needs of the lab. It is expected that Postdoctoral Fellow will engage in both application/disease driven research as well as technical development of new MR pulse sequences, reconstruction methods or image post-processing techniques. The qualifications of a successful candidate should include • A PhD degree in Electrical Engineering, Biomedical Engineering, Physics, Mathematics, Neuroscience or other related fields. • Strong background in signal processing, image processing and quantitative analysis. Excellent proficiency with Matlab and/or Python programming. Skills in C/C++ programming language is a plus. • Demonstrated independent research capabilities and good organizational and inter-personal skills. Previous experience with medical image processing is desirable; it is a strong plus but not necessary to have existing experiences with MRI scanner operation and pulse sequence programming on any of Siemens, GE or Phillips platforms. Postdoctoral fellows at Emory University Department of Radiology also receive an additional fund for professional development activities. Please send your CV and arrange for two letters of recommendation be sent to Dr. Deqiang Qiu (deqiang.qiu(a)emory.edu) and note in the subject line “Postdoctoral Fellow Position”.

1 0

Hippocampal subfields
by Erik O'Hanlon 30 Aug '21

30 Aug '21

External Email - Use Caution Dear FS experts, I've run a lognitudinal hippocampal and amygdala analysis. I jave managed to extract the amygdala and the head body tail subfields bui I'm wondering how to extract the regular FS60 volume measures? I have the three versions (FS60, HBT and CA mgz files in my long mri directories but I just keep getting the HBT volumes. I tried quantifyHAsubregions.sh hippoSf T1.long Outputfilename and that gives me the HBT measures What command do I use to pull out the FS60 subfield measures from my long mri directories? Thanks for any help. Erik

2 4

FreeSurfer qcache Inquiry
by Junyeon Won 28 Aug '21

28 Aug '21

External Email - Use Caution Dear FreeSurfer experts, I am trying to administer group-level analyses using the Human Connectome Project dataset. I downloaded extended processed structural data from the HCP webpage and processed FreeSurfer data were available there. So I ran qcache as a preparatory step for group level analysis but there was an error message (below). ERROR: cannot find /Volumes/JW/FS/100206/surf/lh.white.K The HCP's processed FreeSurfer data did not have lh.white.H/lh.white.K/rh.white.H/rh.white.K files. Could you please let me know if there is any way I can run qcache without the K (or H) files? Or should I run recon-all -qcache my own using the HCP's T1 data? Thank you in advance for your help and have a good weekend. Best, JW ᐧ

2 2

Problem, recon-all never finishes (it gets stuck)
by Samuel Paré 27 Aug '21

27 Aug '21

External Email - Use Caution Hi, I have been trying to process 7T images but for most of them, recon-all just never finishes. It seems like it just bugged at some point with no error messages. At first, I thought that it just needed more time, but once, I had them running for more than 5 days and they were still stuck (they all get stuck at different stages of the process). For example, one get stuck at this stage: After retessellation of defect 13 (v0=21576), euler #=-176 (455542,1350444,894726) : difference with theory (-214) = -38 CORRECTING DEFECT 14 (vertices=77947, convex hull=8486, v0=30684) I have been using Freesurfer and recon-all before and this is the first time it has happened to me. Does someone have an idea how to solve this issue? I am using a macbook pro with 2.6 GHz 6-Core Intel Core i7 and 64GB of RAM. Catalina 10.15.7 Thank you very much, Samuel Paré -- *Samuel Paré, MSc* Laboratoire du développement et plasticité du système visuel École d'optométrie Université de Montréal

3 2

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