Freesurfer April 2024

freesurfer@nmr.mgh.harvard.edu

54 participants
64 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

Retinotopy error computing contrast matrices
by Mesquita, Tiago 14 May '24

14 May '24

External Email - Use Caution Hello FreeSurfer Developers, I am trying to do a retinotopy analysis. My retinotopy has 8 bars, this bars swipe the visual field from the 8 possible directions (all multiple of 45 degrees angle). One complete swipe takes 24 sec. I have 16 different positions per swipe. So I have tried using -retinotopy 24 (period of a complete swipe) or 1.5 (period of each individual bar) in every case i always end up having the same error. For some reason Xcond = NaN (normalized). I tried eccen pos, eccen neg even polar neg and pos in my paradigm files and get always the same result. Can someone help me? this is what i get: Creating Design Matrix ... creation time = 0.006 sec DoMCFit = 1 ntptot = 119, nX = 35, DOF = 84 Saving X matrix to ~/Project/Sess02/bold/rtopy.self.lh/pr001/Xtmp.mat XCond = NaN (normalized) Warning: Matrix is singular to working precision. > In fast_selxavg3b (line 403) Computing contrast matrices Warning: Matrix is singular to working precision. > In fast_selxavg3b (line 441) Warning: Matrix is singular to working precision. > In fast_selxavg3b (line 441) OLS Beta Pass run 1 t= 0.0 reading data ... 0.820008 Global Mean 146.10 Global In-Mask Mean = 146.104 (NaN) Rescale Target = 100 RescaleFactor = 0.684444 OLS Residual Pass run 1 t= 0.0 reading data ... 0.776954 Saving rho1 Found 0 voxels with corrected AR1 > 0.90 Not Whitening Warning: Matrix is singular to working precision. > In fast_glmfit (line 54) In fast_selxavg3b (line 999) Found 7921 zero-valued voxels Computing contrasts Starting contrasts eccen J=2 ------------- Warning: Matrix is singular to working precision. > In fast_fratiow (line 86) In fast_selxavg3b (line 1161) Warning: Matrix is singular, close to singular or badly scaled. Results may be inaccurate. RCOND = NaN. > In fast_fratiow (line 86) In fast_selxavg3b (line 1161) Error using betainc X must be in the interval [0,1]. Error in FTest (line 51) p = betainc(z, dof2/2, dof1/2); Error in fast_selxavg3b (line 1163) pmat = FTest(dof1, dof2, Fmat); If i look into the X.mat m betainc is filled witn NaN. I have atteched my analysis info and X.mat. Let me know if you need more informations about my analysis. Best, Tiago Mesquita

1 1

recon-all process still not finishing properly
by Dong, Yilei 08 May '24

08 May '24

External Email - Use Caution Hi Freesurfer Support, The past few weeks, I've emailed about my recon-all processing stream never completely finishing. We are still encountering the same problem. For context, our recon-all job submission script runs recon-all for 1 image. Given a folder of 100 MRI images, we have another script that calls upon the recon-all script for each image within the folder via a for loop. The result is 100 jobs running in parallel on the cluster. I have attached a screenshot of the parameters we set whenever we submit each recon-all job for each image by SLURM to UCSD's cluster. The maximum time we are allowed for each job is 48 hours. Our jobs are shared-node jobs, which means we run more than 1 job on a single node. This time, we increased RAM from 8GB to 16GB in hopes the entire recon-all processing stream can fully run through for each image, but it still stops at "mri_pretess done" and does not go on to the -fill step. Each job submitted to the cluster took around 4 hours and 20 minutes each to run in our most recent attempt. If increasing the RAM did not change anything for us, how else can we get more verbose error messages? What other reasons could be why our recon-all aborts before finishing? I have attached a recon-all.log from one of our subjects for reference and a screenshot of the parameters we set for submitting jobs in our cluster for reference. Freesurfer version: 7.2.0, but already available as module on UCSD's cluster system Platform: Rocky Linux release 8.8 (Green Obsidian) uname -a: Linux login01 4.18.0-477.15.1.el8_8.x86_64 #1 SMP Wed Jun 28 15:04:18 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux Recon-all log: see attached Thank you! Sincerely, Yilei

2 4

Understanding registration with mri_synthmorph
by Antoine Royer 05 May '24

05 May '24

External Email - Use Caution Hi all! I am working on implementing some freesurfer synth tools into a larger diffusion processing pipeline. Doing so, I would like to use mri_synthmorph to generate a transform and then pass it downstream. FIrstly, I use the dockerized synthmorph version, which seems to present discrepancies in its behaviour vs local testing: 1. Can someone explain the expected output format of freesurfer transforms? When used within docker, I must define affine transforms as .lta and deform transforms as .mgz, otherwise: "ValueError: cannot determine file format from extension for test__init_warp.txt"; while when using mri_synthform from my local freesurfer installation, it doesn't seem like an issue. While I expected .m3z to be the expected output. 2. Here is the sequence of functions I use. My objective is to pass down the resulting transform (affine initialization + deform warp) downstream my pipeline for further registration using ANTs (antsApplyTransforms). I tested and it seems to work, but i would appreciate an opinion. Does the --insrcgeom parameter fixes everything for me? * Initialization + registration (freesurfer): * mri_synthmorph -m affine -t <init.lta> [moving] [fixed] mri_synthmorph -m deform -t <freesurfer_warp.mgz> -o <warped.nii.gz> -i <init.txt> [moving] [fixed] Convert warp from freesurfer (RAS) to ANTs (LPS) using freesurfer warp_convert: mri_warp_convert -g [moving] --inras <freesurfer_warp.mgz> --outlps <ants_warp.nii.gz> Thank you very much!

3 6

Lingering questions for Longitudinal spatiotemporal LME analysis
by Dan Levitas 02 May '24

02 May '24

External Email - Use Caution Hello, I've been following the FreeSurfer longitudinal LME analysis tutorial ( https://secure-web.cisco.com/1UXjrD54pEj2u6JH6Y-Z-dHoSgY6qcbXqPg43uVU2GGyBv…) which has been excellent; however I have a few lingering questions pertaining to my specific analysis. For reference, I am interested in assessing rates of longitudinal atrophy in each of my two groups (0=controls, 1=clinical). Regarding what I've currently done: 1). My Qdec file contains 4 columns: fsid, fsid_base, time(years), group 2). I followed the steps to create the appropriate M matrix, and since my preliminary analysis is simply to create a *group x time* linear model, my design matrix is X = [ones(length(M),1) M M(:,1).*M(:,2)]; and my contrast is CM = [0 0 0 1] 3). I've followed the pipeline for the lme_mass_fit_EMinit, lme_mass_RgGrow, and lme_mass_fit_Rgw functions. 4). I'm skipping the model comparison steps, as I prefer my model with the two random effects (intercept, time). 5). Get stats (F_lhstats = lme_mass_F(lhstats, CM), applied multiple comparison correction ([detvtx,sided_pval,pth] = lme_mass_FDR2(F_lhstats.pval,F_lhstats.sgn,lhcortex,0.05,0); and saved the sided p-values. My questions are: a). Since I'm interested in both hemispheres and the tutorial only assesses the left, do I need to do step #3 with the right hemisphere data? b). Would I need to do the last step in the tutorial to get a single threshold for both hemispheres and then perform the analysis this way? P = [ F_lhstats.pval(lhcortex) F_rhstats.pval(rhcortex) ]; G = [ F_lhstats.sgn(lhcortex) F_rhstats.sgn(rhcortex) ]; [detvtx,sided_pval,pth] = lme_mass_FDR2(P,G,[],0.05,0); pcor = -log10(pth) Thank you for the clarification. Dan

2 1

FW: Infant Freesufer auto-recon failure
by Lemley, Rachel 02 May '24

02 May '24

External Email - Use Caution Good afternoon, We are attempting to run freesurfer on infant data, but are running into an issue with the autorecon-pial code. We corrected the pial surface on the brainmask.mgz in freeview, saved the brainmask.mgz edits, and are trying to run autorecon-pial, but we get the following error that I have attached below. It is looking for a file in surf labeled lh.orig.premesh, but it does not exist. Any suggestions on how to fix this? Am I missing a step? Respectfully, Rachel Lemley The error log is attached below: Subject Stamp: freesurfer-linux-centos7_x86_64-infant-dev-4a14499-20210109-4a14499 Current Stamp: freesurfer-linux-centos7_x86_64-infant-dev-4a14499-20210109-4a14499 INFO: SUBJECTS_DIR is /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects Actual FREESURFER_HOME /n/sw/ncf/apps/infant-freesurfer/4a14499/centos7_x64 Linux holy7c090602.rc.fas.harvard.edu 4.18.0-425.10.1.el8_7.x86_64 #1 SMP Thu Jan 12 16:32:13 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux '/n/sw/ncf/apps/infant-freesurfer/4a14499/centos7_x64/bin/recon-all' -> '/ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02/scripts/recon-all.local-copy' /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02/mri/transforms /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02 /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02 #-------------------------------------------- #@# Mask BFS Mon Apr 22 14:06:18 EDT 2024 /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02/mri mri_mask -T 5 brain.mgz brainmask.mgz brain.finalsurfs.mgz threshold mask volume at 5 DoAbs = 0 Found 401939 voxels in mask (pct= 2.40) MRImask(): AllowDiffGeom = 1 Writing masked volume to brain.finalsurfs.mgz...done. @#@FSTIME 2024:04:22:14:06:19 mri_mask N 5 e 2.09 S 0.01 U 0.76 P 37% M 77200 F 0 R 3233 W 0 c 64 w 51 I 768 O 768 L 2.29 1.57 1.12 @#@FSLOADPOST 2024:04:22:14:06:21 mri_mask N 5 2.29 1.57 1.12 #-------------------------------------------- #@# AutoDetGWStats lh Mon Apr 22 14:06:21 EDT 2024 cd /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02/mri mris_autodet_gwstats --o ../surf/autodet.gw.stats.lh.dat --i brain.finalsurfs.mgz --wm wm.mgz --surf ../surf/lh.orig.premesh infant-dev-4a14499 cd /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02/mri setenv SUBJECTS_DIR /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects mris_autodet_gwstats --o ../surf/autodet.gw.stats.lh.dat --i brain.finalsurfs.mgz --wm wm.mgz --surf ../surf/lh.orig.premesh Reading in intensity volume brain.finalsurfs.mgz Reading in wm volume wm.mgz Reading in surf ../surf/lh.orig.premesh error: No such file or directory error: MRISread(../surf/lh.orig.premesh): could not open file Command exited with non-zero status 1 @#@FSTIME 2024:04:22:14:06:21 mris_autodet_gwstats N 8 e 0.73 S 0.00 U 0.28 P 39% M 43208 F 0 R 1873 W 0 c 23 w 20 I 120 O 0 L 2.29 1.57 1.12 @#@FSLOADPOST 2024:04:22:14:06:22 mris_autodet_gwstats N 8 2.29 1.57 1.12 Linux holy7c090602.rc.fas.harvard.edu 4.18.0-425.10.1.el8_7.x86_64 #1 SMP Thu Jan 12 16:32:13 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux recon-all -s KAN1_02 exited with ERRORS at Mon Apr 22 14:06:23 EDT 2024 For more details, see the log file /ncf/sba08/Triarchy_OA/Rachel/infantFS/subjects/KAN1_02/scripts/recon-all.log To report a problem, see http://secure-web.cisco.com/1hDoMTcotBNpi6ZDqlu1xK1AeLrzrk_QC8e4t7bga0WOa9h…

2 2

Re: [Freesurfer] question about infantfreesurfer
by Knut J Bjuland 02 May '24

02 May '24

External Email - Use Caution What I mean is that it's a beta version, so I just wanted to make sure that I can use these results in a publication. Knut Jørgen Bjuland

2 1

Figures for publication
by Asuka Toyofuku 02 May '24

02 May '24

External Email - Use Caution Dear Freesurfer experts, Thanks to all the help from this community, we're finally done with GLM analysis(mri_glimfit) and the cluster-wise correction (mri_glmfit_sim), and now I'm making figures for publication. My question is 1. Which output mgh files are commonly used for making figures for publication? Do people use, for example, cache.th40.abs.sig.cluster.mgh (cluster-wise corrected map: overlay) or cache.th40.abs.sig.masked.mgh (uncorrected sig values masked by the clusters that survive correction) or even uncorrected ones, such as sig.mgh file in the contrast directory before running the correction? From my impression, not many people use e.g., cache.th40.abs.sig.cluster.mgh (which shows no colour gradation inside the clusters)... 2. How do you select the colour-bar cut-off and max values? Most publications I saw use min cut-off value: -log10(p) = 2.00 and max: -log10(p) = 5.00, but some uses min: p=0.05 and max: p=0.01. Obviously, if you set the max value (with the configure section) very low (e.g., 2.0), the size of the cluster gets bigger and looks more significant. Are there any suggestions for setting the configure/threshold for the figures? Many thanks, Asuka

2 1

[FreeSurfer] recon-all exited with error - watershed error
by Krystal Xiwing Yau 01 May '24

01 May '24

External Email - Use Caution Dear FS experts, I am encountering an error when running recon-all. I followed the FAQ <https://secure-web.cisco.com/1UCt6p3edZmVlY3vv2VBSKmRgX9p8Xmur_qX9QMrgPbk8p…> page to rerun the command but still not running through. Would you please kindly advise? The log file is attached to this email. Thank you. Regards, Krystal Tel.: +852 2255 5024 Fax: +852 2872 5828 Address: Room 304, 3/F, New Clinical Building, 102 Pokfulam Road, Hong Kong.

1 1

Error: Label FG1.mpm.vpnl does not exist in SUBJECTS_DIR fsaverage!
by Pamela González Méndez 25 Apr '24

25 Apr '24

External Email - Use Caution Hi everyone, While running a FreeSurfer command, I encountered the following error message: "Label FG1.mpm.vpnl does not exist in SUBJECTS_DIR fsaverage!" This error seems to indicate that the specified label is missing in the fsaverage directory within the SUBJECTS_DIR. Any insights on resolving this issue would be appreciated. Thank you!

2 1

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer April 2024