Freesurfer

freesurfer@nmr.mgh.harvard.edu

21 participants
30071 discussions

Inquiry Regarding Intrinsic Cortical Curvature (ICC) Calculation in FreeSurfer
by vittal korann 23 Jul '25

23 Jul '25

External Email - Use Caution Dear FreeSurfer Team, I’m currently working on a project involving cortical morphology and am particularly interested in calculating Intrinsic Cortical Curvature (ICC) using FreeSurfer. I came across a recent publication ( https://secure-web.cisco.com/1oL8IZPRfQ5o5UDqVHfF8rU4SLPTHeXEdS4OHOftgKkxaX…) that utilized ICC in their statistical analysis. I’d like to clarify a couple of points regarding how ICC is derived within the FreeSurfer framework: 1. Is Gaussian curvature as computed by FreeSurfer considered equivalent to ICC? 2. If not, are there additional steps or specific tools required to compute ICC beyond the standard curvature outputs (e.g., via mris_curvature or mris_curvature_stats)? Any guidance or documentation you could share would be greatly appreciated. I’m aiming to ensure methodological consistency with published work and would be grateful for your insights. Thank you for your time and support. With regards, Vittal

2 1

`recon-all -base` fails with glibc/license error, but cross-sectional `recon-all` works
by Foyzul Rahman 23 Jul '25

23 Jul '25

External Email - Use Caution Hi FreeSurfer community, I'm running into an issue with the longitudinal FreeSurfer pipeline. Background: I'm working with T1-weighted MRI scans from an aerobic exercise intervention study in older adults (age 60+). I have two timepoints per subject: pre- and post-intervention. Cross-sectional `recon-all` has successfully completed for both timepoints. I'm now trying to run the longitudinal pipeline using `recon-all -base`, but I'm getting a glibc/license-related error. Command: ```bash recon-all -base sub002base \ -tp /rds/projects/s/segaertk-chbh-00077/Structural/FreeSurfer/sub002_tp1 \ -tp /rds/projects/s/segaertk-chbh-00077/Structural/FreeSurfer/sub002_tp2 \ -all ``` Error: ``` GNU libc version: 2.28 ERROR: Systems running GNU glibc version greater than 2.15 require a newly formatted license file (it's free). Please download a new one from the following page: http://secure-web.cisco.com/1S-XRYZwdfIh6LBaYDmryClXzVhMGn3s7IZR6N9Pa3Ony7_… ``` Notes: I have a valid, updated license file (downloaded from the official site). I set the license path before running the command: ```bash export FS_LICENSE=${HOME}/license.txt ``` Weirdly, the cross-sectional `recon-all` works fine using the same environment and license file. I'm using FreeSurfer version: FreeSurfer/6.0.1-centos6_x86_64 My system is running glibc 2.28. What I've Tried: * Verified that `$FS_LICENSE` is exported and correct. * Checked for any outdated `license.txt` in `$FREESURFER_HOME` and replaced it. * Confirmed that `recon-all -version` and `which recon-all` return the expected FreeSurfer install. * Deleted partial `sub002base` folder and retried clean. Has anyone encountered this behaviour - where the cross-sectional `recon-all` works, but `recon-all -base` fails due to the license/glibc issue? Any help would be greatly appreciated! Warmest regards, Foyzul. Dr Foyzul Rahman Centre for Human Brain Health (CHBH) University of Birmingham United Kingdom

1 0

FreeSurfer 8.1.0 released (includes important bug fix!)
by Douglas N. Greve 22 Jul '25

22 Jul '25

We are pleased to release freesurfer 8.1.0. This contains an important bug fix so we recommend that you this instead of 8.0.0. It will give the same output as 8.0.0 while avoiding the "symlink" bug described below. The biggest new feature is the new infant freesurfer pipeline. Congrats to Lilla Zoelli, Ashvath Balgovind, and Jackson Nolan for getting this completed!. More details below and at https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes * New features: infant_recon, dcm2niix upgrade, a few speed ups, gtmseg uses samseg by default, better support for ANTS N3, mideface works better with multiple frame, vlrmerge-mni, mri_gtmpvc can create PMOD compatible csv files, added -skip-nu-intensity-cor to recon-all, * Bug fixes: o Fixes recon-all where it would create a symlink from orig.mgz to nu.mgz when run a second time. This could happen if you edited your data and then rerun recon-all. o All the things from the patch (csvprint, xhemi, mri_segstats, -nonuintensitycor). o Fixed --lesion-mask-pattern in samseg script.

1 0

TRACULA statistics - len_center issue
by Gabriella Monopoli 21 Jul '25

21 Jul '25

External Email - Use Caution Dear experts, I analyzed a few subjects from my dataset using TRACULA (FreeSurfer version 8). While examining the pathstats.overall.txt file for the left anterior longitudinal fasciculus, I noticed an anomaly: the Len_Center value is greater than Len_max. This occurs for all the subjects analyzed and also appears in other tracts. I was wondering whether this issue might be related to my data, a bug in the anlysis, or perhaps it's not actually an error at all. I'm sure you'll be able to help me understand how to resolve this and properly analyze my data. I've attached, as example, the pathstats.overall.txt file of one of my subjects. Best regards, Gabriella

1 0

2D vs 3D gradient distortion correction (mri_gradunwarp)
by Annchen Knodt 20 Jul '25

20 Jul '25

External Email - Use Caution I have a longitudinal dataset collected on a Siemens Skyra, where online distortion correction on the T1s was turned OFF for the first wave several years ago, and with an upgrade to XA30 since then, it is now turned ON for the second wave of data collection and performing a 2D (in-plane) correction. Obviously, we want the acquisitions to match as closely as possible between the two timepoints, so I am trying to figure out how to use gradunwarp to perform post-hoc correction on the first scans. A problem I am running into (see my thread on HCP’s gradunwarp github page<https://secure-web.cisco.com/1UEsGaoLMJInc49tXx1leDHidHoj0m6ydT6viNz9VH1gN5…>) is that the software performs a 3D correction, which is not the same as the 2D correction being applied to my wave 2 data. I don’t yet know of a way to either: 1. apply just a 2D correction to my wave 1 uncorrected data to match what the scanner is doing at wave 2 (I’ve communicated with a Siemens rep who says there is no documentation he can share on the 2D correction to help me figure out how to do this) or 2. UNDO the 2D correction on my wave 2 data, so that I can then use gradunwarp to apply the 3D correction to both waves. I don’t have a very good understanding of the differences between the 2D and 3D corrections, and I do not currently have a dataset with both uncorrected and 2D-corrected copies of the same scan to allow me to experiment with and test various strategies (hoping to get this soon, but not sure when I’ll be able to). As such, I am hoping that someone with more experience or knowledge on this might be able to shed some light on the nature of the issue and severity of the problem and/or offer guidance on how I might be able to correct it. Thanks! Annchen

2 1

Re: Freesurfer output. Segmentation fault (core dumped) during parcellation step
by fsbuild 18 Jul '25

18 Jul '25

External Email - Use Caution This posting indicates that changing the path to ./stats from ../stats got rid of a segmentation fault.  I would at least check that all paths in the arguments ro mris_anatomical_stats are correct though. https://secure-web.cisco.com/17sKLSeDVET5-5S3ZEgRGL6lX55gFUMqV2UWISNObj-rC9… But I don’t see the error in the recon -all log you sent which is apparently from May 2020 showing Freesurfer 6.0.0 is being run on RedHat Enterprise Linux 7.5. - R. On Jul 10, 2025, at 16:24, Samantha Hendry <shendry2(a)ed.ac.uk> wrote:mris_anatomical_stats -th3 -mgz -cortex ../label/lh.cortex.label -f ../stats/lh.aparc.stats -b -a ../label/lh.aparc.annot -c ../label/aparc.annot.ctab FMS_0094 lh white

3 3

FreeSurfer license request
by 李昊宇 18 Jul '25

18 Jul '25

External Email - Use Caution Dear FreeSurfer team, I would like to obtain a FreeSurfer license for academic research use. Due to the registration page can't run properly and gives the 'FreeSurfer Registration Error reCAPTCHA not checked' error message, I wrote this email to request for a license file. Here are the required details: First Name: Haoyu Last Name: Li Institution: Shanghai Jiao Tong University Type of institution which will use this software: Non-profit education/research email: haoyu_li(a)sjtu.edu.cn Telephone: +86 19817863375 Operating System/Platform: Linux Number of Users: 1 Please send the license file to this email address. Thank you very much! Best regards, Haoyu Li

2 1

Fwd: Make Docker of FreeSurfer Dev version.
by fsbuild 18 Jul '25

18 Jul '25

External Email - Use Caution We do not support end users building freesurfer from source.   Development builds can be unstable and may not work and/or work the same way depending upon when you download them; the same is true for checking out the source code on any given day. You should instead create a docker container running a standard Ubuntu 22 release, e.g., LTS 22.0.4, for which there already exists a .deb installer package corresponding to a stable freesurfer release.   The freesurfer releases are listed here with the latest release being 8.0.0, https://secure-web.cisco.com/1xCL7oeqhtHSjdQttn-ynKQW05sIF78iIeQcw7h6OTSotE… Install a freesurfer Ubuntu 22 .deb package in the container using the standard install command, e.g., if you download the 8.0.0 release installer,https://secure-web.cisco.com/16l9n1feB3EbaBIKS8nXkmTUTR-uy0sofOYG… then the install command would be,$ sudo apt install freesurfer_ubuntu22-8.0.0_amd64.deb - R.

1 0

Re: FreeSurfer license request
by fsbuild 18 Jul '25

18 Jul '25

External Email - Use Caution There is likely some javascript not loading in your browser preventing something like captcha graphics from displaying.  Check to make sure the browser is not blocking any javascript and/or try a different browser.   Part of filling out the form and registering includes using the mouse or pointing device to click on and accept the terms of the displayed license agreement.   We do not distribute licenses via email since that would bypass the legal consent step. - R. On Jul 18, 2025, at 03:28, 李昊宇 <haoyu_li(a)sjtu.edu.cn> wrote:       External Email - Use Caution         Dear FreeSurfer team, I would like to obtain a FreeSurfer license for academic research use. Due to the registration page can't run properly and gives the 'FreeSurfer Registration Error reCAPTCHA not checked' error message, I wrote this email to request for a license file. Here are the required details: First Name: Haoyu Last Name: Li Institution: Shanghai Jiao Tong University Type of institution which will use this software: Non-profit education/research email: haoyu_li(a)sjtu.edu.cn Telephone: +86 19817863375 Operating System/Platform: Linux Number of Users: 1 Please send the license file to this email address. Thank you very much! Best regards, Haoyu Li _______________________________________________ Freesurfer mailing list -- freesurfer(a)nmr.mgh.harvard.edu To unsubscribe send an email to freesurfer-leave(a)nmr.mgh.harvard.edu https://secure-web.cisco.com/1n8TEBGY47Kdny8w5LiDfbzot98S-MbfB_DPUMKIAPfRl_…

1 0

Make Docker of FreeSurfer Dev version.
by Óscar Peña-Nogales 18 Jul '25

18 Jul '25

External Email - Use Caution Dear, I'm attempting to create a *docker version of the FreeSurfer Development version* following the steps shown in https://secure-web.cisco.com/1zahxQetlgW4KiuOIq4HdcILTBQvomXB-frywITaD7T90r… Nevertheless, I'm encountering some issues, probably related to how I set up the docker, but they are blocking me. I attach the Dockerfile I'm trying to make (up to the step where it fails). The error raised by the make command is: *[ 57%] Building CXX object mris_convert/CMakeFiles/mris_convert.dir/mris_convert.cpp.oIn file included from /usr/include/ITK-5.2/gifti_io.h:4, from /root/freesurfer/include/gifti.h:26, from /root/freesurfer/mris_convert/mris_convert.cpp:40:/usr/include/ITK-5.2/itk_zlib.h:33: error: "ITK_USE_SYSTEM_ZLIB" redefined [-Werror] 33 | #define ITK_USE_SYSTEM_ZLIB | <command-line>: note: this is the location of the previous definitioncc1plus: all warnings being treated as errorsmake[2]: *** [mris_convert/CMakeFiles/mris_convert.dir/build.make:80: mris_convert/CMakeFiles/mris_convert.dir/mris_convert.cpp.o] Error 1make[1]: *** [CMakeFiles/Makefile2:10396: mris_convert/CMakeFiles/mris_convert.dir/all] Error 2make: *** [Makefile:146: all] Error 2* Last, see I'm interested in making the docker base image in Ubuntu 22.04. Thanks in advance, Óscar

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