Freesurfer

freesurfer@nmr.mgh.harvard.edu

11 participants
30059 discussions

Re: [Freesurfer] Bug reporting - longmc: case-sensitive issue in variable
by fsbuild 19 Mar '24

19 Mar '24

External Email - Use Caution  On Ubuntu linux the default shell may still be “dash” and not the “bash” shell.  Dash cannot parse all bash scripting, so type “bash” or try setting “bash” to be your default shell and then run commands. - R. On Mar 19, 2024, at 22:17, Boscutti, Andrea <Andrea.Boscutti(a)uth.tmc.edu> wrote:        External Email - Use Caution        Hello FreeSurfer Developers,When attempting to run "recon-all -long" in subjects with multiple T1 inputs, I get the following error when "longmc" is run:concatLtas: Undefined variable.if: Expression Syntax.Command exited with non-zero status 1For the entire log, see attached "longmc_error.log".The following change at line 188 of "freesurfer/bin/longmc" fixes it (case-sensitive issue in "$concatLtas"):set ud = `UpdateNeeded $origvol $CrossList $concatLtas  -->  set ud = `UpdateNeeded $origvol $CrossList $ConcatLtas`Best,Andrea<longmc_error.log>_______________________________________________Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

2 1

Measure statistics from manually segmented cortical GM
by K Eswar 19 Mar '24

19 Mar '24

External Email - Use Caution Dear Freesurfer Team, My name is K Eswar, and I am reaching out to seek assistance regarding a project I am currently working on. I have manually segmented the cortical gray matter of a fetal brain MRI (nii.gz and stl formats) and now need to perform various analyses such as measuring curvature, volume, area, and thickness. As I am relatively new to using FreeSurfer, I am seeking guidance on how to effectively utilize the software for these analyses. Could you please provide me with some guidance or resources on how to proceed with this task using FreeSurfer? Any assistance or advice you could offer would be greatly appreciated. Thank you very much for your time and assistance. I look forward to hearing back from you soon. Kind regards K Eswar

1 0

Issues with accuracy of pial and white matter surface reconstruction in infant_recon_all command
by Tsigaras, Thanos 19 Mar '24

19 Mar '24

External Email - Use Caution Dear infantFS-team, We are a team of PhD students, who recently decided to use infantFS for the segmentation and surface extraction on our data. We used the infant_recon_all command for preprocessing T1w sMRI scans from the Baby Connectome Project dataset. The pipeline runs without any errors, but we encountered noticeable inaccuracies in the reconstructed surfaces, especially concerning the pial surface. The white matter surface reconstruction was mostly accurate (with some exceptions), but in some cases even the segmentation fails as well, particularly in very young subjects. Even when the segmentation is accurate, the reconstruction of the pial surface fails to align correctly with the grey matter boundaries, often (if not always) in the medial temporal lobes. All scans have undergone quality control to include only those with minimal motion artefacts. The age range of the subjects analysed, spans from 0 to 4 years old. All scans were realigned and intensity-corrected using the iBEAT toolbox. We would gladly appreciate any assistance in understanding the causes of this problem and are looking forward to methods or changes that we need to follow in order to solve them. Attached you can find the script we used for the analysis of the scans in a .txt form. Please feel free to ask for any further information, details or materials regarding this issue. Thanks a lot in advance for your time and help! Best regards Thanos Tsigaras ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Forschungszentrum Jülich GmbH 52425 Jülich Sitz der Gesellschaft: Jülich Eingetragen im Handelsregister des Amtsgerichts Düren Nr. HR B 3498 Vorsitzender des Aufsichtsrats: MinDir Stefan Müller Geschäftsführung: Prof. Dr. Astrid Lambrecht (Vorsitzende), Karsten Beneke (stellv. Vorsitzender), Dr. Ir. Pieter Jansens ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------

1 0

Recon-all command produces only 001.mgz file in the mri/orig directory
by Naama Zur 17 Mar '24

17 Mar '24

External Email - Use Caution Hi all, I am using Ubuntu 23.1 with freesurfer 7.4.1 to run the recon-all command and the command runs without error and I do get a set of directories including: label, mri, scripts, stats, surf, tmp, touch, and trash. However, the only folders that are not empty are scripts where there are 11 files and a 001.mgz file in the mri/orig directory. I am not sure what I am doing incorrectly and was hoping for some guidance with this. Thank you!

2 1

Re: [Freesurfer] Large discrepancy in left and right volumes of perirhinal/entorhinal cortices on FreeSurfer 7.1
by Douglas N. Greve 14 Mar '24

14 Mar '24

I've looked into this a little bit. I looked at the area of the original manual labels for entorhinal as found in the aparc. They do show a lh>rh asym of 3%, but it is not significant (p=.45). I looked at the automatic labels in two other large data sets. I found significant lh>rh asym of ento area in both of about 8-10%. I don't consider 10% asym in area to be a symptom of something problematic. This would represent an expansion or contraction of the entorhinal border of about 0.5mm. I'm guessing that this is beyond the accuracy that we can expect. I was looking at surface area, but now I see that you are looking at volume. The manual labels were defined on an areal basis, and I assume that that was the basis of the volume asym, but it could be due to differences in thickness. How big a difference are you seeing in volume? Can you look at area and thickness to see if you see differences there? Also, the surfaces are really hard to place in this area. Oftentimes, the ento will have a big bite in it that has to be fixed manually. I don't think there is asym in the bites, but it could make a particular case look extreme. In general, we don't recommend doing laterality studies with the aparc as none of the steps involved have been developed to make sure that laterality artifacts are removed. In this case, it sounds like you are checking laterality for QA purposes? On 3/13/2024 9:30 PM, Katie Vandeloo wrote: > External Email - Use Caution > > Hi Dr. Greve, > > I hope you are doing well. I am hoping to get in touch with someone from MGH about some issues with Freesurfer 7. Our group hasn’t yet been successful in finding the right person to talk to about this! > > We noticed that there are large discrepancies between left and right hemisphere volumes of the perirhinal cortex (exceptionally small volumes in the right hemisphere) in the outputs of recon-all for Freesurfer 7 but NOT Freesurfer 6. I have reached out to other labs who observed the same pattern. I think there is something systemic about Freesurfer 7 that should be looked into. Just wanted to make someone aware - please let me know if there is someone else I should contact about this. > > Thank you for your time! > > Kind Regards, > Katie Vandeloo

1 0

GCS file format specification and conversion to GIFTI
by Oscar Esteban 14 Mar '24

14 Mar '24

External Email - Use Caution Dear all, I'm trying to understand the GCS file format as I failed to try an easy conversion with mris_convert. If mris_convert cannot convert GCS to GIFTI, is there any open description of how the format is implemented? Thanks very much Best, Oscar *___________________________* *Oscar Esteban, Ph.D.* Research and Teaching Ambizione FNS Fellow Dept. of Radiology, CHUV, University of Lausanne +41 77 957 00 77 +1 (650) 733 33 82

2 1

mri_label2vol fillribbon issue
by Schneider, Letitia Maria 14 Mar '24

14 Mar '24

External Email - Use Caution Hello, I am using Freesurfer version 7.3.2 on a mac. I am using the mri_label2vol function to project an anatomical label into volume space, but I realised the label comes out wrong when I use the --fill-ribbon flag: its orientation changes to RSP. The same label comes out right when I use the --fillthresh flag (oriented in RPI). Any idea why using --fillribbon changes the orientation? Thank you in advance. Best wishes, Letitia

2 1

recon-all hard error: mris_calc could not establish read access to rh.area
by Ariko, Taylor 13 Mar '24

13 Mar '24

External Email - Use Caution Hello FreeSurfer Developers, I am attempting to run the recon-all pipeline on one subject using the following code: recon-all -all -i t1.nii -s p03_1 Although, I get the following error in the command line: mris_calc -o rh.area.mid rh.area add rh.area.pial error: No such file or directory error: read_signa(/Volumes/Beta/Carmen/t1/p03_8/p03_8/surf/I.001): could not open file error: No such file or directory error: mris_calc: could not establish read access to 'rh.area'. Tue Aug 1 15:58:13 EDT 2023 ERROR: vertexvol mris_calc -o rh.area.mid rh.area add rh.area.pial Darwin C02F10PVPN5W.local 22.6.0 Darwin Kernel Version 22.6.0: Wed Jul 5 22:21:56 PDT 2023; root:xnu-8796.141.3~6/RELEASE_X86_64 x86_64 recon-all -s p03_8 exited with ERRORS at Tue Aug 1 15:58:13 EDT 2023 To report a problem, see http://secure-web.cisco.com/1dFtz9tfZWpoyZVj7mXFMLhzADOYxytPQ-lAzdm9SS0VD1F… The recon-all.error file contains: ------------------------------ SUBJECT p03_8 DATE Tue Aug 1 15:58:13 EDT 2023 USER taylorariko HOST C02F10PVPN5W.local PROCESSOR x86_64 OS Darwin Darwin C02F10PVPN5W.local 22.6.0 Darwin Kernel Version 22.6.0: Wed Jul 5 22:21:56 PDT 2023; root:xnu-8796.141.3~6/RELEASE_X86_64 x86_64 7.2.0 (freesurfer-darwin-macOS-7.2.0-20210713-aa8f76b) /Applications/freesurfer/7.2.0/bin/recon-all PWD /Volumes/Beta/Carmen/t1/p03_8/p03_8/mri CMD vertexvol --s p03_8 --rh --th3 I have made sure that I have read, write, and execute access to the folder /path/to/subject/surf and the file rh.area. I can successfully run the command to make rh.area (mris_place_surface --area-map ../surf/rh.white ../surf/rh.area), but the error persists in using rh.area. I have disk space and can run other commands such as this fake one (mris_calc -o rh.area.Test rh.area.pial add rh.area.pial), but not if it contains rh.area. I have tried to rerun the command from root and get the same error message. I found 2 similar threads of the mail-archive, though I didn’t find any remedies: 1. https://secure-web.cisco.com/1s_8L81vMyC0-KdGicsKrUniSkh4Egg5_LxXz-_YnmsgL0… 2. https://secure-web.cisco.com/1MW6Y468pIeNLaB3ruN22mzrkJIjqalKQmsNncuNqyZW9S… Does anyone have any thoughts on how to trouble-shoot this? I have included my recon-all.log in the case it is of any use. 1. Freesurfer version: freesurfer-darwin-macOS-7.2.0-20210713-aa8f76b 2. Platform: macOS Sonoma 14.3.1 3. uname -a: Darwin C02F10PVPN5W.local 23.3.0 Darwin Kernel Version 23.3.0: Wed Dec 20 21:28:58 PST 2023; root:xnu-10002.81.5~7/RELEASE_X86_64 x86_64 4. recon-all.log: see attached Thanks in advance! Taylor Ariko

3 10

Unsubscribe from the mailing list
by Tina Zahrie 13 Mar '24

13 Mar '24

External Email - Use Caution Hi, I would like to unsubscribe from the mailing list. Thanks

1 0

[WARNING: ATTACHMENT UNSCANNED]mri_ca_normalize in recon-all -base generate blank norm.mgz
by Keshuo Lin 13 Mar '24

13 Mar '24

External Email - Use Caution Hello FreeSurfer Developers, I am attempting to use Freesurfer 7.4.1 recon-all-clinical and longitudinal process for stroke study. 38 subjects out of 85 subjects failed to complete the longitudinal process. The error is mri_cc -aseg aseg.auto_noCCseg.mgz -o aseg.auto.mgz -lta /g/data/ey6/keshuo/strokog/sss/freesurfer_result/threetime/000216RRW_template/mri/transforms/cc_up.lta 000216RRW_template error: mri_cc: no WM voxels found with norm > 40 -- check skull stripping I checked the previous steps between successful subjects and unsuccessful subjects. The norm.mgz generated from mri_ca_normalize are different. The subject_template/mri/norm.mgz of unsuccessful subjects are blank. The issue is because no control point was detected. Can you tell me how to solve this issue? I randomly attached one of the successful subjects and one of the unsuccessful subjects norm.mgz brainmask.mgz and norm_template.mgz. I have also attached the error recon-all log. Platform: I run the above codes in two OS systems: One is "Rocky Linux" VERSION="8.9 (Green Obsidian)" The other is Mac M2 chips with MACOS venture 13.3.1 Thanks for your time and patience. Best Regards, Keshuo Lin

1 0

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