Freesurfer November 2025

freesurfer@nmr.mgh.harvard.edu

32 participants
40 discussions

Clarification on running -autorecon2-wm, -autorecon-pial, and -autorecon3 after manual white matter and pial edits
by Camilo Castelblanco Riveros 05 Jan '26

05 Jan '26

External Email - Use Caution Hello FreeSurfer Developers and Community, I am currently refining surface reconstructions for my subjects by performing manual edits on both the white matter segmentation and pial surfaces using FreeSurfer. According to the FreeSurfer documentation, after making white matter edits, the recommended command is: recon-all -autorecon2-wm -subjid <subject> and after pial surface edits, it is recommended to run: recon-all -autorecon-pial -subjid <subject> followed by: recon-all -autorecon3 -subjid <subject> My main goal is to ensure that both white matter and pial manual edits are correctly incorporated without losing either. However, I have seen in the FreeSurfer tutorial the combined command: recon-all -autorecon2-wm -autorecon3 -subjid <subject> which runs both stages sequentially. I have two related questions: 1. If I run `recon-all -autorecon2-wm -autorecon3` after both white matter and pial edits, will my manual pial edits be preserved, or will they be overwritten by the `-autorecon2-wm` step? 2. Is the recommended workflow to run `-autorecon2-wm` first, then `-autorecon-pial` as a separate command, and finally `-autorecon3`? Or can some of these stages be combined safely in a single command line? For reference, I am using: - FreeSurfer version: 7.4.1 - Platform: Red Hat Enterprise Linux With gratitude, Camilo =================================== *Camilo Castelblanco Riveros (He/His)* Neuroscience Ph.D. Candidate (G4) Robertson Lab <https://secure-web.cisco.com/1VGQeQdcjwLtBi3C9UBo55c7SqVai-UrjQm1lF9QaAVlL0…> & Bujarski Lab <https://secure-web.cisco.com/1y_9Zgr30ubJ5JlJFc0I6-yJcgPm8sSkph5TvzTpqhl6nv…> Student Interest Group in Neurology (SIGN <https://secure-web.cisco.com/1uGgfMp5acPV-NG-koj4ZIyk0_wqPKAvcoF_BCuwGLKcw_…>) Co-Leader Global Health Fellow e: ccastelblanco.gr(a)dartmouth.edu Zoom <https://secure-web.cisco.com/1i_IGOhLxqGYKgaY8rs_7Ga3801HHvx4C0j37X5CbmatR6…>

2 1

Recon failing: normal vector of length zero at vertex NNNNN with N faces
by Sambamoorthy, Sowmya 23 Dec '25

23 Dec '25

External Email - Use Caution Hello! Need help getting past these errors (attached recon log) for reference. Many of my brains are throwing errors like below: "normal vector of length zero at vertex 101440 with 6 faces first tangent vector of length zero at vertex 101440" Last few lines of recon.log below: [cid:image001.png@01DC27E0.966D3810] Any help appreciated! Thank you in advance for your support. --Sowmya

3 6

GLM on runs with different conditions
by Xu, Yaoda 23 Dec '25

23 Dec '25

External Email - Use Caution Hi there, I have a large number of conditions to include in a study, like 60. So I split the conditions into multiple runs, with 40 conditions per run, say run 1 with conditions 1 to 40, run 2 with conditions 21 to 60, and run 3 with conditions 1 to 20 and 41 to 60. Can I put all three runs in a run list and run a single GLM with 60 conditions? Normally, I would have the same number of conditions in each run when I put a bunch of runs together in a single GLM to increase SNR. In this case, different runs would share some of the conditions, but not all conditions appear in each run. I just want to make sure that it is OK to run a GLM like this and get a valid beta estimate for each of the 60 conditions. Thanks a lot. Yaoda Xu

2 1

make_average_subject
by Don Hagler 23 Dec '25

23 Dec '25

External Email - Use Caution We have used make_average_subject to resample the white and pial surfaces of an individual subject to ico 7 to produce surface coordinates in the standard tessellation, allowing for group analyses of cortical shape. It may be worth noting that we temporarily replace the talairach.xfm in the subject’s mri/transforms directory with an identify matrix so that the resampled surfaces have the same position and shape as the native surfaces but with the standard ico 7 tesselation. This resampling procedure works great with FreeSurfer version 5.3, but as of version 6.0 or 7.1.1, when a single subject is supplied to make_average_subject, it runs for a while but then eventually produces an error like this: nframes = 1 Computing std/var across frames ERROR: cannot compute std from one frame ERROR: make_average_surface While we can continue doing this resampling step using FS 5.3 (e.g., after running the initial recon with 7.1.1), we would like to stop relying on this very old version. Is there an alternative approach to re-tesellation of an individual subject’s surfaces into ico? Or could a future release of FreeSurfer include a modification to make_average_surface to allow it to once again handle a single subject as input?

2 2

PETSurfer v8.1.0 vs 7.x.x — missing pons in gtmseg.ctab and rescale issue in mri_gtmpvc
by Carme Uribe 23 Dec '25

23 Dec '25

External Email - Use Caution Hi all, I’m running into an issue after upgrading from FreeSurfer 7.1.2 to 8.1.0 when using PETSurfer. In v7.1.2, the gtmseg.ctab included the *pons (Id 174)*, which allowed the subsequent mri_gtmpvc step to perform the default *rescale normalization based on pons intensity*. However, in v8.1.0, the pons no longer appears in the gtmseg.ctab. Since the rescale step still expects that region by default, the pipeline fails when trying to normalize by the pons. From the release notes I see that gtmseg now uses SAMSEG by default, and I suspect that this change may have modified how subcortical or brainstem structures are handled. Also, any chance the gtmseg step could add other regions typically used as reference (e.g., wm/whole cerebellum, centrum semiovale) maintaining the aseg and aparc parcellations from FS? Thanks very much in advance for your help, Carme

2 1

Re: freesurfer license error: qdec not recognize license
by fsbuild 23 Dec '25

23 Dec '25

External Email - Use Caution   Freesurfer 6  does not use the FS_LICENSE environment variable to find the license, e.g., by setting it to the path of the license.txt file in your home directory.  You can try making a copy of license.txt to be read as $FREESURFER_HOME/.license You should NEVER post the contents of a license file to a public online forum. - R.

3 2

expert.opts is not parsed correctly in FS 8
by Billah, Tashrif 23 Dec '25

23 Dec '25

https://surfer.nmr.mgh.harvard.edu/fswiki/recon-all#ExpertPreferences-1 Did you guys change how parameters from expert.opts is ingested by recon-all in FS 8.1.0? We have successfully been using the following line in expert.opts file with FS 7: mris_inflate -n 30 (https://surfer.nmr.mgh.harvard.edu/fswiki/mris_inflate) But now, FS 8.1.0 recon-all gives an error: ``` mris_inflate -no-save-sulc 30 ../surf/rh.smoothwm.nofix ../surf/rh.inflated.nofix Not saving sulc Reading 30 ** failed to open GIFTI XML file '30.gii' mrisReadGIFTIdanum: ERROR reading gifti 30.gii [INFO] read, cannot find 30, trying 30.gii ... error: No such file or directory error: mris_inflate: could not read surface file 30 ``` This almost certainly means FS 8.1.0 is not parsing expert.opts line correctly. The "-n 30" is bein parsed as just "30". Best, Tashrif

3 4

Support regarding QDEC
by Henry Do 23 Dec '25

23 Dec '25

External Email - Use Caution Hi freesurfer experts, I am working on a project, and I have used mri_glmfit to compare the cortical thickness of the patient cohort with that of the control. When I examined the results, I noticed some unexpected clusters of atrophy and wanted to investigate which subjects contributed to each cluster. In the previous version, I could do this easily with QDEC, but unfortunately, my current computer can’t set up QDEC because the Windows version is not compatible with FreeSurfer 6.0. I hope you can help me with an alternative method for performing this analysis, as it would be greatly beneficial for my project. Sincerely, Henry Hiệp ĐỖ (Henry DO) Tran Minh Hiep DO (Henry DO) PhD Candidate | B.Biomed.Eng Faculty of Science, School of Psychology THE UNIVERSITY OF SYDNEY FRONTIER Frontotemporal Dementia Research Group | Brain and Mind Centre | M02G | Level 1, 94 Mallet Street | Camperdown | NSW | 2050 Contact: +61 451 973 339 Email: trdo0838(a)sydney.edu.au<mailto:trdo0838@sydney.edu.au>

2 1

wm edits, fs8.0.0: rerunning leaves wm-surfs unchanged
by Zahnert, Felix 23 Dec '25

23 Dec '25

Dear experts, I have run recon-all using Freesurfer 8.0.0 and corrected wm.mgz in some subjects (deleting voxels where ?h.white grossly extended into gm by assigning them values of 1; adding voxels where juxtacortical hypointensities were misclassified as gm by assigning them 255). After these edits (and saving wm.mgz), I ran recon-all -s <subject> -autorecon2-wm -autorecon3. However, this left the reconstructed surfaces virtually unchanged. Even areas of larger edits are left unchanged. I did check for small areas of stray voxels that could have led to such misclassification. In the documentation, it is recommended to run recon-all with the -make flag in such cases, which no longer seem to be supported. Do you have any recommendations? Thank you so much, Best, Felix

3 7

DTI Registration issue
by Ramesh Babu 10 Dec '25

10 Dec '25

External Email - Use Caution Hello, I am analyzing my DTI scans in Ubuntu 22 LTS with FreeSurfer 8.1. I am facing a registration issue when I run the following script. # Setting Up Environment BASE_DIR="/mnt/data_analysis/freesurfer_data/sr_dti_analysis/subj02" SUBJECTS_DIR="$BASE_DIR" SUBJECT="subj02_fs" export SUBJECTS_DIR SUBJECT # Define Input and Output Subfolders T1_DICOM="$BASE_DIR/t1_SPGR" DTI_DICOM="$BASE_DIR/epi_dti" T1_NIFTI="$BASE_DIR/t1_nifti" DTI_NIFTI="$BASE_DIR/dti_nifti" DTI_DIR="$BASE_DIR/$SUBJECT/dtrecon" MRI_DIR="${SUBJECTS_DIR}/${SUBJECT}/mri" OUTDIR="$DTI_DIR" # Resampling subject's wmparc, aparc+aseg, and brain volumes to diffusion space for vol in wmparc aparc+aseg brain; do vol_file="${SUBJECTS_DIR}/${SUBJECT}/mri/${vol}.mgz" vol2diff="${OUTDIR}/${vol}2diff.mgz" cmd=(mri_vol2vol --mov "${OUTDIR}/lowb.nii" \ --targ "$vol_file" \ --inv \ --interp nearest \ --o "$vol2diff" \ --reg "${OUTDIR}/register.dat" \ --no-save-reg) echo "${cmd[@]}" "${cmd[@]}" done # Mask FA, ADC, ICV volumes using wmparc2diff.mgz for vol in fa adc ivc; do cmd=(mri_mask "${OUTDIR}/${vol}.nii" \ "${SUBJECTS_DIR}/${SUBJECT}/mri/wmparc2diff.mgz" \ "${OUTDIR}/${vol}-masked.mgz") echo "${cmd[@]}" "${cmd[@]}" done # Resampling masked volumes to MNI space interp="trilin" template="$FREESURFER_HOME/average/mni152/mri/brain.mgz" # MNI template morph="${SUBJECTS_DIR}/${SUBJECT}/mri/transforms/${SUBJECT}_to_mni.m3z" # Example morph for vol in fa adc ivc; do vol_file="${OUTDIR}/${vol}-masked.mgz" outvol="${OUTDIR}/${vol}-masked_MNI.mgz" cmd=(mri_vol2vol --targ "$template" \ --m3z "$morph" \ --noDefM3zPath \ --reg "${OUTDIR}/register.dat" \ --mov "$vol_file" \ --o "$outvol" \ --interp "$interp" \ --no-save-reg) echo "${cmd[@]}" "${cmd[@]}" done Error: INFO: This REGISTER_DAT transform is valid only for volumes between COR types with c_(r,a,s) = 0. /mnt/data_analysis/freesurfer_data/sr_dti_analysis/subj02/subj02_fs/dtrecon/lowb.nii.gz /mnt/data_analysis/freesurfer_data/sr_dti_analysis/subj02/subj02_fs/mri/wmparc.mgz error: File exists error: LTAchangeType: src geometry must be valid Then I did the BB registration separately. bbregister --s subj02_fs \ --mov $DTI_DIR/lowb.nii.gz \ --dti \ --init-fsl \ --reg $DTI_DIR/register.dat Still, I am facing the same issue. Please provide some solutions. Best Ramesh

2 5

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer November 2025