Freesurfer July 2024

freesurfer@nmr.mgh.harvard.edu

44 participants
44 discussions

Re: [Freesurfer] Question on using thalseg_1.1.h5
by Georgios Argyropoulos 23 May '25

23 May '25

External Email - Use Caution Many thanks and apologies, Henry - I figured it out... I stupidly replaced your improved segmentation script in both the scripts and the bin directory instead of only replacing the former. many thanks for your time, much appreciated g

2 5

run_samseg: Optimizer: maximalDeformation is too small; stopping; killed
by Ringki Marie Siewert 08 Aug '24

08 Aug '24

2 2

Longitudinal analyses using Repeated measures anova
by Josefine Tingdal Taube Danielsen 05 Aug '24

05 Aug '24

External Email - Use Caution Hi Freesurfer support, I'm using Freesurfer version 7.4.1 and Ubuntu 22_x86_64. freesurfer-linux-ubuntu22_x86_64-7.4.1-20230614-7eb8460. I'm trying to run longitudinal analysis using Repeated Measures Anova (https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova) My study has 2 groups (MM=patient group and HC=healthy controls) and 2 time points (1 and 2). I am looking at cortical thickness and cortical volume. Here is some information: We have been using these Contrast files: # Create 1-vs-2 contrast file echo "0 0 1" > 1-vs-2.mtx # Create group-effect contrast file echo "0 1 0" > group-effect.mtx # Create interaction contrast file echo "0 1 -1" > interaction.mtx # Create mean contrast file echo "1 0 0" > mean.mtxfree And then followed this: Preproc using FSGD mris_preproc --target fsaverage --hemi lh --meas thickness --out lh.thickness.mgh --fsgd rmanova.fsgd Smooth data mri_surf2surf --s fsaverage --hemi lh --fwhm 5 --sval lh.thickness.mgh --tval lh.thickness.sm05.mgh Run GLM fit mri_glmfit --glmdir lh.rmanova --y lh.thickness.sm05.mgh --fsgd rmanova.fsgd doss \ --C 1-vs-2.mtx --C group-effect.mtx --C interaction.mtx --C mean.mtx \ --surface fsaverage lh This was our results: INFO: gd2mtx_method is doss Reading source surface /mnt/c/CogimmunStudy/FS/fsaverage/surf/lh.white Number of vertices 163842 Number of faces 327680 Total area 65417.000000 AvgVtxArea 0.399269 AvgVtxDist 0.721953 StdVtxDist 0.195472 7.4.1 cwd /mnt/c/CogimmunStudy/FS cmdline mri_glmfit --glmdir lh.rmanova --y lh.thickness.sm05.mgh --fsgd rmanova.fsgd doss --C 1-vs-2.mtx --C group-effect.mtx --C interaction.mtx --C mean.mtx --surface fsaverage lh sysname Linux hostname D56979 machine x86_64 user josdan FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /mnt/c/CogimmunStudy/FS/lh.thickness.sm05.mgh logyflag 0 usedti 0 FSGD rmanova.fsgd labelmask /mnt/c/CogimmunStudy/FS/fsaverage/label/lh.cortex.label maskinv 0 glmdir lh.rmanova IllCondOK 0 ReScaleX 1 DoFFx 0 SigUseDouble 1 Creating output directory lh.rmanova Loading y from /mnt/c/CogimmunStudy/FS/lh.thickness.sm05.mgh ... done reading. INFO: gd2mtx_method is doss Saving design matrix to lh.rmanova/Xg.dat Computing normalized matrix Normalized matrix condition is 1 Matrix condition is 3.15 Found 149955 points in label. Pruning voxels by thr: 1.175494e-38 Found 149955 voxels in mask Saving mask to lh.rmanova/mask.mgh Reshaping mriglm->mask... search space = 74612.577841 DOF = 123 Starting fit and test Fit completed in 0.0220333 minutes Computing spatial AR1 on surface Residual: ar1mn=0.986733, ar1std=0.003725, gstd=3.501824, fwhm=8.246166 Writing results 1-vs-2 maxvox sig=-2.06583 F=7.13254 at index 102162 0 0 seed=1720706961 Computing efficiency group-effect maxvox sig=114.013 F=8295.3 at index 155689 0 0 seed=1720706961 Computing efficiency interaction maxvox sig=106.775 F=6297.15 at index 64378 0 0 seed=1720706961 Computing efficiency mean maxvox sig=133.337 F=17231.6 at index 155689 0 0 seed=1720706961 Computing efficiency mri_glmfit done Now our questions: 1. Have we been using the right contrasts? 2. Do the results look realistic? 3. Is it possible to have some more information about results (e.g. summary, areas being significant, visualization, sig.mgh, t-values, p-values not i -log10)? Thank you! Best, Josefine Danielsen

3 5

Interpreting LME results with multi-row contrasts
by Dan Levitas 05 Aug '24

05 Aug '24

External Email - Use Caution I am performing a longitudinal LME analysis in which participants were scanned anywhere from 2-6 times over the course of several years. There are 4 groups denoting participants: Controls (C, coded as 0), group 1 (G1, coded as 1), group 2, (G2, coded as 2), and group 3 (G3, coded as 3). Additionally, I have included two covariates (age and number of previous concussions). With a single random effect (intercept, all 1s), my design matrix is: *Yij = intercept + time + age + numPrevConcussion + G1 + time*G1 +G2 + time*G2 + G3 + time*G3* I am primarily interested in time*group interaction differences between G1 and G2 (G1 - G2) and G1 and G3 (G1 - G3). I first tried single row contrasts separately (i.e. CM.C = [0 0 0 0 0 1 0 -1 0 0] for G1-G3, and CM.C = [0 0 0 0 0 1 0 0 0 -1] for G1 - G3), which produced no significant results. I then created a multi-row contrast of the two contrasts above (i.e. CM.C = [zeros(2,5) [-1 0 1 0 0;-1 0 0 0 1]]), which also did not reveal significant findings. Only when I created a 3-row contrast (CM.C = [zeros(3,5) [1 0 0 0 0;1 0 -1 0 0;1 0 0 0 -1]]) did I get significant results. I'm a bit unsure how to properly interpret these contrasts and wanted to get clarification on my thinking. Is the 3-row contrast assessing the null hypothesis of no group differences in the rate of change over time among the four groups? If so, does the lack of findings in the 2-row and single row contrasts indicate that there are no interaction differences between G1 and G2, and between G1 and G3? Thank you

2 1

FreeSurfer in fmriprep
by Tengwen Fan 01 Aug '24

01 Aug '24

External Email - Use Caution Dear FreeSurfer Developers, I hope this email finds you well! I am using fMRIprep to do pre-processing. The fMRIprep combines methodology from AFNI, ANTs, Freesurfer, FSL, and Mindboggle to do preprocessing. An error was reported while running the fMRIprep. It seems that it was due to the structural data and it was about the registration by FreeSurfer. But I am new in the neuroimaging area and not very sure about what the error actually is about. I asked the fmriprep team in neurostars (https://neurostars.org/t/fmriprep-error-makeisotropic-warning-not-different…) and they are also not sure about this and recommended to ask the FreeSurfer. Although I am not sure if the FreeSurfer in fmriprep is what you also work with, I would like to ask and would really appreciate it if you could help explain what error it is. Thanks you so much! The error is attached below. Please let me know if you have any questions or if more information is needed! [https://neurostars.org/uploads/default/original/2X/1/1898e24a8821a606fad2dd…]<https://neurostars.org/t/fmriprep-error-makeisotropic-warning-not-different…> Fmriprep error: makeIsotropic WARNING: not different enough, won't reslice!<https://neurostars.org/t/fmriprep-error-makeisotropic-warning-not-different…> Summary of what happened: Hi, I am running fmriprep and getting the following error. I would really appreciate any help! Please let me know if more information should be provided! Thanks so much, tengwen Command used (and if a helper script was used, a link to the helper script or the command generated): fmriprep-docker /Users/Desktop/WL_fMRI/data/BIDS /Users/Desktop/WL_fMRI/data/derivatives/fmriprep participant --participant-label sub-WLfMRI035 --fs-license-file /Users/Desktop/WL_fMRI/licen... neurostars.org The Error is: Cmdline: mri_robust_register --satit --iscale --lta /tmp/work/fmriprep_23_2_wf/sub_WLfMRI033_wf/anat_fit_wf/surface_recon_wf/fsnative2t1w_xfm/T1_robustreg.lta --mov /out/sourcedata/freesurfer/sub-WLfMRI033/mri/T1.mgz --dst /tmp/work/fmriprep_23_2_wf/sub_WLfMRI033_wf/anat_fit_wf/anat_template_wf/denoise/mapflow/_denoise0/sub-WLfMRI033_ses-Session1_T1w_noise_corrected.nii.gz Stdout: 7.3.2 --satit: Will iterate with different SAT to ensure outliers below wlimit! --iscale: Enabling intensity scaling! --lta: Output transform as /tmp/work/fmriprep_23_2_wf/sub_WLfMRI033_wf/anat_fit_wf/surface_recon_wf/fsnative2t1w_xfm/T1_robustreg.lta . --mov: Using /out/sourcedata/freesurfer/sub-WLfMRI033/mri/T1.mgz as movable/source volume. --dst: Using /tmp/work/fmriprep_23_2_wf/sub_WLfMRI033_wf/anat_fit_wf/anat_template_wf/denoise/mapflow/_denoise0/sub-WLfMRI033_ses-Session1_T1w_noise_corrected.nii.gz as target volume. reading source '/out/sourcedata/freesurfer/sub-WLfMRI033/mri/T1.mgz'... reading target '/tmp/work/fmriprep_23_2_wf/sub_WLfMRI033_wf/anat_fit_wf/anat_template_wf/denoise/mapflow/_denoise0/sub-WLfMRI033_ses-Session1_T1w_noise_corrected.nii.gz'... Registration::setSourceAndTarget(MRI s, MRI t, keeptype = TRUE ) Type Source : 0 Type Target : 3 ensure both FLOAT (3) Reordering axes in mov to better fit dst... ( -1 -3 2 ) Determinant after swap : 1 Mov: (0.5, 0.5, 0.5)mm and dim (512, 512, 512) Dst: (0.5, 0.5, 0.5)mm and dim (360, 512, 512) Asserting both images: 0.5mm isotropic - reslicing Mov ... -- changing data type from 0 to 3 (noscale = 0)... - reslicing Dst ... -- Original : (0.5, 0.5, 0.5)mm and (360, 512, 512) voxels. -- Resampled: (0.5, 0.5, 0.5)mm and (512, 512, 512) voxels. -- Reslicing using cubic bspline MRItoBSpline degree 3 - WARNING: trg mri outside_val = 0 but suspected background: -0.00687514 If background is black or darkgray, ignore this. If white, pass --whitebgdst Registration::findSaturation - computing centroids - computing initial transform -- using translation info - Get Gaussian Pyramid Limits ( min size: 16 max size: -1 ) - Build Gaussian Pyramid ( Limits min steps: 0 max steps: 4 ) - Build Gaussian Pyramid ( Limits min steps: 0 max steps: 4 ) - Max Resolution used: 3 -- gpS ( 64 , 64 , 64 ) -- gpT ( 64 , 64 , 64 ) - running loop to estimate saturation parameter: -- Iteration: 1 trying sat: 16 min sat: 0 ( -1 ), max sat: 16 ( 0.12101 ), sat diff: 16, (wlimit=0.16) -- Iteration: 2 trying sat: 8 min sat: 8 ( 0.324589 ), max sat: 16 ( 0.12101 ), sat diff: 8, (wlimit=0.16) -- Iteration: 3 trying sat: 12 min sat: 12 ( 0.193726 ), max sat: 16 ( 0.12101 ), sat diff: 4, (wlimit=0.16) -- Iteration: 4 trying sat: 14 min sat: 12 ( 0.193726 ), max sat: 14 ( 0.15173 ), sat diff: 2, (wlimit=0.16) -- Iteration: 5 trying sat: 13 min sat: 13 ( 0.17074 ), max sat: 14 ( 0.15173 ), sat diff: 1, (wlimit=0.16) -- Iteration: 6 trying sat: 13.5 - final SAT: 14 ( it: 6 , weight check 0.15173 <= 0.16 ) Registration::computeMultiresRegistration - computing centroids - computing initial transform -- using translation info - Get Gaussian Pyramid Limits ( min size: 16 max size: -1 ) - initial transform: Ti = [ ... 1.0000000000000 0 0 1.6982521922295 0 1.0000000000000 0 -1.1595074345649 0 0 1.0000000000000 -1.1415599519338 0 0 0 1.0000000000000 ] - initial iscale: Ii =1 Resolution: 4 S( 32 32 32 ) T( 32 32 32 ) Iteration(f): 1 -- intensity log diff: abs(-0.85599) -- diff. to prev. transform: 1.62699 Iteration(f): 2 -- intensity log diff: abs(-0.0720915) -- diff. to prev. transform: 0.761125 Iteration(f): 3 -- intensity log diff: abs(-0.00194135) -- diff. to prev. transform: 0.0383579 Iteration(f): 4 -- intensity log diff: abs(-0.00123869) -- diff. to prev. transform: 0.0128975 Iteration(f): 5 -- intensity log diff: abs(-0.000750416) <= 0.001 :-) -- diff. to prev. transform: 0.00518321 <= 0.01 :-) Resolution: 3 S( 64 64 64 ) T( 64 64 64 ) Iteration(f): 1 -- intensity log diff: abs(0.00702447) -- diff. to prev. transform: 0.977321 Iteration(f): 2 -- intensity log diff: abs(9.94659e-05) <= 0.001 :-) -- diff. to prev. transform: 0.0258922 Iteration(f): 3 -- intensity log diff: abs(1.21569e-05) <= 0.001 :-) -- diff. to prev. transform: 0.00163623 <= 0.01 :-) Resolution: 2 S( 128 128 128 ) T( 128 128 128 ) Iteration(f): 1 -- intensity log diff: abs(-0.000545933) <= 0.001 :-) -- diff. to prev. transform: 0.356881 Iteration(f): 2 -- intensity log diff: abs(-7.53945e-05) <= 0.001 :-) -- diff. to prev. transform: 0.0881187 Iteration(f): 3 -- intensity log diff: abs(-5.00529e-05) <= 0.001 :-) -- diff. to prev. transform: 0.0253314 Iteration(f): 4 -- intensity log diff: abs(-2.0764e-05) <= 0.001 :-) -- diff. to prev. transform: 0.00744419 <= 0.01 :-) Resolution: 1 S( 256 256 256 ) T( 256 256 256 ) Iteration(f): 1 -- intensity log diff: abs(0.00127177) -- diff. to prev. transform: 0.268975 Iteration(f): 2 -- intensity log diff: abs(0.000348898) <= 0.001 :-) -- diff. to prev. transform: 0.0666025 Iteration(f): 3 -- intensity log diff: abs(8.4156e-05) <= 0.001 :-) -- diff. to prev. transform: 0.016444 Iteration(f): 4 -- intensity log diff: abs(2.77798e-05) <= 0.001 :-) -- diff. to prev. transform: 0.00492133 <= 0.01 :-) Resolution: 0 S( 512 512 512 ) T( 512 512 512 ) Iteration(f): 1 -- WARNING: mem usage large: 4668.08Mb mem + 6 MRI Maybe use --subsample <int> Stderr: makeIsotropic WARNING: not different enough, won't reslice! Killed Traceback: Traceback (most recent call last): File "/opt/conda/envs/fmriprep/lib/python3.10/site-packages/nipype/interfaces/base/core.py", line 398, in run runtime = self._post_run_hook(runtime) File "/opt/conda/envs/fmriprep/lib/python3.10/site-packages/niworkflows/interfaces/freesurfer.py", line 248, in _post_run_hook fix_lta_length(lta_file) File "/opt/conda/envs/fmriprep/lib/python3.10/site-packages/niworkflows/interfaces/freesurfer.py", line 399, in fix_lta_length lines = Path(lta_file).read_text().splitlines(keepends=True) File "/opt/conda/envs/fmriprep/lib/python3.10/pathlib.py", line 1134, in read_text with self.open(mode='r', encoding=encoding, errors=errors) as f: File "/opt/conda/envs/fmriprep/lib/python3.10/pathlib.py", line 1119, in open return self._accessor.open(self, mode, buffering, encoding, errors, FileNotFoundError: [Errno 2] No such file or directory: '/tmp/work/fmriprep_23_2_wf/sub_WLfMRI033_wf/anat_fit_wf/surface_recon_wf/fsnative2t1w_xfm/T1_robustreg.lta' Best, Tengwen

2 4

misalignment after converting to gifti
by Jiaen Liu 31 Jul '24

31 Jul '24

External Email - Use Caution I converted surface data in the cortex to surf.gii and loaded the data in connectom workbench. But the cortical curvature aligned well with the T1 image in Freesurfer doesn't align well in the workbench. I used mris_convert for the conversion. I'm wondering if I missed certain parameter Below the left is the workbench display and right is in Freesurfer. Thanks! [cid:image001.png@01DAE359.488E22C0] Best regards, Jiaen Liu Ph.D. Assistant Professor Advanced Imaging Research Center & Radiology UT Southwestern Medical Center 2201 Inwood Ln Dallas, TX, USA, 75390 Office: 214.645.2750 Website: Liu (Jiaen) Lab | UT Southwestern, Dallas, Texas<https://labs.utsouthwestern.edu/liu-jiaen-lab> ________________________________ UT Southwestern Medical Center The future of medicine, today.

2 2

Question: Recon-all Using T1/T2 Mutual Information
by Edwards, Vincent 31 Jul '24

31 Jul '24

External Email - Use Caution Hello Freesurfer team! We are editing brain scans using T1/T2 mutual information to improve the original image. Are the recon-all flags the same with as they would be when editing using only the T1 images (i.e., recon-all -cp, -pial, -wm), or are there additional flags are needed after editing? Thank you, Vincent Edwards

2 1

Data Analyst Position at University of Miami Miller School of Medicine
by Otero Ospina, Andrea 30 Jul '24

30 Jul '24

External Email - Use Caution Hi I hope this mail finds you well. Could you please post a job announcement to your mailing list? I'm already inscribed in your mailing list. Best regards Data Analyst Position at University of Miami Miller School of Medicine The Gabbay Laboratory, located within the Department of Psychiatry & Behavioral Sciences at the University of Miami Miller School of Medicine, is seeking well-qualified applicants for the role of data analyst. This position will focus on processing human neuroimaging datasets, including anatomical and functional magnetic resonance imaging (MRI), MR spectroscopy, and related biological measures. Candidates should hold an MS or equivalent degree in the fields of computer science, neuroscience, psychology, radiology, engineering, biostatistics, or a related field. Candidates with a BS-level degree and significant prior experience in a similar role will also be considered on a case-by-case basis. Proficiency with one or more major programming languages used in bioscience (i.e. MATLAB, Python, R) as well as Unix/Linux environments is expected. Familiarity with neuroimaging software (e.g. FSL, AFNI, SPM, Freesurfer, Caret, Connectome Workbench) and with scientific/cluster computing is preferred. The primary responsibility for this role will be to analyze neuroimaging data from multiple ongoing and recently concluded research projects; additional tasks will entail assisting with MRI data collection and manuscript drafting. The data analyst position will be supervised by Vilma Gabbay, MD and Benjamin Ely, PhD. Our lab is dedicated to the study and treatment of mood, anxiety, and substance use disorders in high-risk populations (e.g. adolescents, people living with HIV, cannabis users) using innovative neuroimaging and immunological methodologies. The neuroimaging program features high-resolution data acquisition and analysis techniques modeled on the Human Connectome Project (HCP) and is committed to the highest standards for magnetic resonance imaging. The lab is funded by four ongoing NIH R01 neuroimaging studies (R01MH120601, R01DA054885, R01MH128878, and R01MH131207) and several smaller grants, providing unparalleled funding for active and exploratory studies. Project abstracts with further details can be found on NIH RePORTER. Resources available on-campus and through long-standing collaborations include multiple research-dedicated 3T MRI scanners with cutting-edge accelerated and multi-echo fMRI sequences; high-performance scientific computing clusters; and an extensive library of analytic and statistical software. The PIs are fully committed to fostering a productive mentorship environment, with opportunities for career advancement including manuscript coauthorship, conference attendance, and training in grant writing. Please send a CV, statement of interest, and contact information for three references to axo833(a)med.miami.edu<mailto:axo833@med.miami.edu>. University of Miami is an equal opportunity employer, and all qualified candidates will be given equal consideration. Priority will be given to candidates available to start sooner. Compensation will be proportional to experience and will follow standard rates posted by University of Miami. Andrea Otero, M.D. Clinical Research Coordinator Department of Psychiatry and Behavioral Sciences University of Miami — Miller School of Medicine Email: axo833(a)med.miami.edu<mailto:njh115@med.miami.edu> Phone: 305-243-7086 [cid:17329a4c-56a0-4992-a963-79be54c54353] [cid:47ee1259-91e2-4e19-b42d-696df19c9a75]<https://outlook.office.com/bookwithme/user/0e91508c58304aa8bcb3ea3526547a26…> Book time to meet with me<https://outlook.office.com/bookwithme/user/0e91508c58304aa8bcb3ea3526547a26…>

1 0

FreeSurfer Question
by Selena Gangaram 29 Jul '24

29 Jul '24

External Email - Use Caution Hi, I hope you are doing well, I subscribed to your mailing list to ask my question! I was wondering if you knew the reason as to why my extraction code in FreeSurfer listed regions such as "Whole_hippocampal_body" amongst other brain regions twice in my stats files. In my excel stats files I have two Whole_hippocampal_body and I wasn't sure what the two different values meant. Please help! Thank you!!! Best, Selena

2 1

Inquiry About Calculation of Integrated Rectified Gaussian Curvature in mris_anatomical_stats
by littlepotato 29 Jul '24

29 Jul '24

External Email - Use Caution Dear FreeSurfer Support Team, I hope this message finds you well. I am currently using FreeSurfer's mris_anatomical_stats tool for my research, and I have encountered a question regarding the calculation of the Integrated Rectified Gaussian Curvature value reported in the .stats files, such as lh.aparc.stats. Specifically, I would like to understand the detailed methodology used to compute the Integrated Rectified Gaussian (Mean) Curvature. My current understanding is that this value might be derived from the mean of the absolute values of the Gaussian curvature multiplied by the vertex areas, similar to the following calculation: mean(abs(lh.smoothwm.K.crv)×lh.area)\text{mean}(\text{abs}( \text{lh.smoothwm.K.crv} ) \times \text{lh.area}) Could you please confirm if this is correct? If not, I would appreciate it if you could provide a detailed explanation of the exact steps and calculations involved in deriving this metric. Thank you very much for your time and assistance. I look forward to your response. Best regards, Amber Lin Researcher Tianjin Medical University lly_scu(a)163.com

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 July 2024