Freesurfer

freesurfer@nmr.mgh.harvard.edu

12 participants
30081 discussions

Re: [Freesurfer] Native volume to fs_average surface
by Glasser, Matthew 20 Apr '21

20 Apr '21

External Email - Use Caution You would do that through the subject’s own surface. It does not work very well to map data directly from the volume onto group average surfaces (see e.g. Coalson et al., 2018 PNAS). Matt. From: <freesurfer-bounces(a)nmr.mgh.harvard.edu> on behalf of Mason Wells <WellsMT1(a)cardiff.ac.uk> Reply-To: Freesurfer support list <freesurfer(a)nmr.mgh.harvard.edu> Date: Monday, April 19, 2021 at 4:23 AM To: Freesurfer support list <freesurfer(a)nmr.mgh.harvard.edu>, "freesurfer-bounces(a)nmr.mgh.harvard.edu" <freesurfer-bounces(a)nmr.mgh.harvard.edu> Subject: [Freesurfer] Native volume to fs_average surface External Email - Use Caution Hi experts, I have some fMRI data that I have pre-processed in Feat. I want to resample a single subject's data to the fs_average surface but unsure of the most optimal way of doing this. At present I am warping the functional data to the MNI brain in the volume (using FSL's applywarp) and then resampling to fs_average using the following command. * mri_vol2surf --src filtered_func_MNI.nii --out filtered_func_surf.mgh --hemi rh --mni152reg My question is whether you think this is the most appropriate approach, and/or if there is way to move directly from native volume to fs_average surface without needing to go via MNI volume? Best wishes, Mason Mason T Wells, MSc PhD student School of Optometry and Vision Sciences & Cardiff University Brain Research Imaging Centre (CUBRIC), School of Psychology Cardiff University Cardiff CF24 4HQ UK Email: WellsMT1(a)cardiff.ac.uk<mailto:WellsMT1@cardiff.ac.uk> Tel: 02920 879628 Web: Cardiff University webpage<https://secure-web.cisco.com/1qGEoQnnU3besQKg-GQwdKq-WtCCzi1b5tl-HNLywYRQ48…> Mason T Wells, MSc Myfyriwr PhD Yr Ysgol Optometreg a Gwyddorau’r Golwg & Canolfan Ymchwil Delweddu’r Ymennydd Prifysgol Caerdydd (CUBRIC), Yr Ysgol Seicoleg Prifysgol Caerdydd Heol Maindy Caerdydd CF24 4HQ DU E-bost: WellsMT1(a)caerdydd.ac.uk<mailto:WellsMT1@caerdydd.ac.uk> Ffôn: 02920 879628 ________________________________ The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

3 9

Building atlas by Xhemi: Segmentation fault (core dumped)
by Li-Zhen Chen 20 Apr '21

20 Apr '21

External Email - Use Caution Hello Freesurfer Developer， I'm building my own symmetric atlas by Xhemi, something is wrong at the final step(Make final iteration) Here is my code: make_average_subject --out myatlas.i3 --surf-reg myatlas.i2.sphere.reg --subjects sub_01 sub_02 --xhemi --lh When I ran this code, a problem emerged. Here is part of the log: mris_seg2annot --seg /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.vote.mgh --ctab /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.1.sub_01.xhemi.ctab --s myatlas.i3 --hemi lh --o /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/lh.aparc.a2009s.annot $Id: mris_seg2annot.c,v 1.10 2014/11/15 00:07:19 greve Exp $ cwd /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3 cmdline mris_seg2annot --seg /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.vote.mgh --ctab /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.1.sub_01.xhemi.ctab --s myatlas.i3 --hemi lh --o /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/lh.aparc.a2009s.annot sysname Linux hostname chenlz-HP-ZHAN-99-Pro-G2-Microtower-PC machine x86_64 user chenlz subject myatlas.i3 hemi lh surfseg /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.vote.mgh ctab /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.1.sub_01.xhemi.ctab annotfile /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/lh.aparc.a2009s.annot Reading surface seg /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.vote.mgh Reading surface /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/surf/lh.white Reading ctab /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.1.sub_01.xhemi.ctab Writing annot to /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/lh.aparc.a2009s.annot Annot2std-Run-Time-Sec 16 annot2std Done mri_annotation2label --subject myatlas.i3 --hemi lh --outdir /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/make_cortex_label.tmp.8214 --annotation aparc.a2009s subject = myatlas.i3 annotation = aparc.a2009s hemi = lh outdir = /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/make_cortex_label.tmp.8214 surface = white Reading surface /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/surf/lh.white Loading annotations from /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/lh.aparc.a2009s.annot reading colortable from annotation file... colortable with 11176 entries read (originally /media/chenlz/Zixuans1/MRI_Data/Human_MRI/CCS_PrePro/myatlas.i3/label/tmpdir.annot2std.8488/seg.1.sub_01.xhemi.ctab) Segmentation fault (core dumped) ERROR: make_average_surface I run Freesurfer 6.0 on Ubuntu 18.04. I've tried to rerun my script several times but I can't figure this problem out. What could be the possible reasons? Any answer will be appreciated! Kind regards, Jewel

2 2

question about lesion segmentation using samseg
by Ritobrato Datta 19 Apr '21

19 Apr '21

External Email - Use Caution Hi All, I have T1w MPRAGE, T2w and PD images but I dont have FLAIR images from a Multiple Sclerosis study. Can I use samseg to segment the lesions ? Can you please suggest ? Many thanks Ri

2 1

Tracula group analysis in freesurfer
by Zeng, Qi 19 Apr '21

19 Apr '21

External Email - Use Caution Hi Experts, I understand there is surface-based group analysis in Freesurfer, which is for thickness, curvature, volume and etc. Is there a group analysis for DTI tracula results? Any links to recommend to conduct the analysis? Thank you so much! Best, Qi

2 1

Abormally high foldind index
by Ellen Ji 19 Apr '21

19 Apr '21

External Email - Use Caution Dear Experts, While the majority of our subjects have a folding index between 5-30, there are a couple of *extreme* outliers in the thousands or even negative. Could this hint that a particular part of recon-all failed and should be rerun? Thank you for your interpretation and advice! -Ellen

2 3

Autorecon1 failure
by Abdollahi, Shervin (NIH/NINDS) [C] 19 Apr '21

19 Apr '21

External Email - Use Caution Hello Freesurfer Developers, am trying to use run freesurfer that is configured inside smriprep.workflow.anatomical.py under init_anat_preproc_wf pipeline. this workflow, does the following: 1. T1w reference: realigning and then averaging T1w images 2. Brain Extraction and INU (bias field) correction (antsBrainExtraction.sh) 3. Brian tissue segmentation (@fsl_fast) 4. Spatial normalization to standard space (antsRegistration) 5. Surface reconstruction with Freesurfer When it gets to step 5, it takes skull-stripped t1, ants_segm, and bias_corrected image to use in freesurfer recon-all command. In the surface reconstruction section, it starts by running the following command: recon-all -autorecon1 -i /path-to /sub-03_ses-20140624_T1w.nii -T2 /path-to/sub-03_ses-20140624_T2w.nii -noskullstrip -hires -openmp 1 -subjid sub-03 -sd /Users/abdollahis2/Desktop/fcd_nipype/derivatives/freesurfer -expert /Users/abdollahis2/Desktop/fcd_nipype/single_subject_03/anat_preproc_wf/surface_recon_wf/_subject_id_03/autorecon1/expert.opts And surely soon, it gives me an error that it can not read mgh file (brainmask.mgz). looking into freesurfer/sub-03/mri directory I can see that brainmask.mgz is not created and I am not sure if this is because the option -noskullstrip was used? I have attached the recon-all.log file. I would appreciate any suggestion on how to debug this problem. 1. Freesurfer version: 7.1.1 (freesurfer-darwin-macOS-7.1.1-20200811-8b40551) 2. Platform: macOS Catalina, Version 10.15.7 3. uname -a: Darwin ndsd-eeg-cedar 19.6.0 Darwin Kernel Version 19.6.0: Tue Jan 12 22:13:05 PST 2021; root:xnu-6153.141.16~1/RELEASE_X86_64 x86_64 4. Recon-all.log: see attached

2 9

Longitudinal processing: QA
by Malo Gaubert 19 Apr '21

19 Apr '21

External Email - Use Caution Hello FreeSurfer developers, I am trying to apply longitudinal processing on T1w MRI data for two timepoints (~8 months between the two TP) in adolescents in order to perform vertex-based analyses. I am following the online tutorial here for the preprocessing: https://secure-web.cisco.com/1AR-A4eBY63Zb7ogV-L0wed0Mf63cN1cf9Evvcu56pwG2m… Thus I launched the creation of BASE followed by the LONGITUDINAL. Everything went well for recon-all for these 3 steps. Now, I want to do the QA on the data and I have two questions regarding this QA: 1/ For one subject, for the two timepoints, the white and pial surface are entering in the left lateral ventricle as displayed in figure 1 below (session 1 of the subject; cross-sectional), while the aseg is fine (this area is considered as Lateral ventricle, see figure 2, see below). What correction could I apply for this case ? 2/ For some subjects, there is a small shift between the pial of the two sessions (or the white; see figure 3 (see below)=display of the norm of my norm and pial/white for the two timepoints (long)) and/or some weird/non straight/zigzag lines for pial or white (see figure 3, especially between temporal pole and the rest of the brain). What could be the problem in this case? Could it be related to some bad quality of T1w images? 1) FreeSurfer version: freesurfer-linux-centos7_x86_64-7.1.0.tar.gz 2) Platform: Ubuntu 18 (Virtual Machine on remote server) 3) uname -a: Linux ubuntu18-cbrain-042020 4.15.0-135-generic #139-Ubuntu SMP Mon Jan 18 17:38:24 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux Any help would be appreciated. Best regards, Malo Link to figure 1: https://secure-web.cisco.com/1Z0OIVNfAyd0P4dX5TvUfticgFxyac5XLFP9qZ15LY2KoH… Link to figure2: https://secure-web.cisco.com/1dkQLGxh5x41ISY3MzCyMUowvQmrn9m-TXAqgZfq29hFI5… Link to figure3: https://secure-web.cisco.com/1Nqbq7hapdkZjMI0hXAMBIcoKgeG0dkh3UzuuOGbjE_sUP…

1 0

Talairach Failed: Distorted T1
by Harkey, Thomas Jarrott 18 Apr '21

18 Apr '21

External Email - Use Caution Hello, My name is Thomas Harkey and I am conducting research using FreeSurfer. After running recon-all on this patient, the Talairach failed according to the .log file. No stats were calculated but loading the orig.mgz file into FreeView looks like this (see image attached). This DICOM file was successfully uploaded into other programs. How can I fix this error? Kind regards, -- Thomas Harkey University of Arkansas for Medical Sciences College of Medicine, Class of 2022 501-553-6318 TJHarkey(a)uams.edu<mailto:TJHarkey@uams.edu> ---------------------------------------------------------------------- Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.

2 9

glmfit with aseg table: view surface overlay
by Griem, Julia 18 Apr '21

18 Apr '21

External Email - Use Caution Dear Freesurfer team, I am comparing subcortical volumes (aseg table) in 2 groups with no other variables. Based on the GLM lecture slides from the course I attended, I ran the following glmfit command: mri_glmfit --table aseg_table.txt --fsgd FSGD/DemographicsSubcortBinaryGroup.fsgd --C Contrasts/HC-ASPD.mtx --C Contrasts/ASPD-HC.mtx --glmdir roi.binarygroup.glmdir It created all the correct output and subdirectories. I then used freeview -f $SUBJECTS_DIR/fsaverage/surf/lh.inflated:overlay=ASPD-HC/sig.mgh but an error saying "failed to load surface overlay" comes up and the terminal error message is "number of vertices in overlay data does not match with surface". I assume this is because I am trying to load a 'whole-brain' sig.mgh onto a lh surface? Do I have to change the glmfit command to specify hemispheres, and if so, how do I do that since I do not want to look at cortical but subcortical data? Or alternatively, can I load both hemispheres into freeview so the overlay sig.mgh fits? Thank you so much, Julia

2 5

Re: [Freesurfer] Running recon-all with FLAIR
by Edina Szabó 18 Apr '21

18 Apr '21

External Email - Use Caution Hi Douglas, Thank you for your response! I really appreciate it. I ran the command – it works if I replace --i with --mov. How can I use the new FLAIRraw.lta to create a new FLAIR-based pial surface? For example, should I run recon-all -autorecon2 or some later step? Also, we were assuming that registration of the FLAIR to the T1 was the problem (underestimated pial surface), since the FLAIR.mgz produced by Freesurfer was not well registered to the T1.mgz. So, if that's the case, and I have another well-registered FLAIR (using FSL or SPM), is there a way to skip the Freesurfer registration and use that FLAIR to create the FLAIR pial surface? Thank you for your help in advance! All the best, Edina Szabo __ Edina Szabo, PhD Research Fellow Department of Anesthesiology, Critical Care and Pain Medicine Boston Children's Hospital | Harvard Medical School ----------------- Try this cd subject/mri mri_coreg --i orig/T2raw.mgz --targ orig.mgz --reg transforms/T2raw.lta You can run with multiple threads by adding --threads N where N is the number of threads On 3/29/2021 5:18 PM, Edina Szabó wrote: External Email - Use Caution Dear All, We are running recon-all on the latest version of FreeSurfer (v7.1.1). We are using both T1 and FLAIR images to better define the pial surfaces, but when we run the recommended command line (recon-all -subject subjectname -i /path/to/input_volume -FLAIR /path/to/FLAIR_volume -FLAIRpial -all) the pial surface is very underestimated (?.pial.FLAIR – and the aseg volume is based on these surfaces), and we get better results without the FLAIR. It seems that FreeSurfer mis-registers the FLAIR to the T1 during the process. Is there a separate step we should do to register FLAIR to T1? Thank you, Edina _______________________________________________ Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer _______________________________________________ Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

2 1

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