Freesurfer January 2023

freesurfer@nmr.mgh.harvard.edu

87 participants
111 discussions

Start a nNew thread

Fwd: mri_robust_template - MyMatrix::RotationMean Error
by Andrew Beaudin 07 Feb '23

07 Feb '23

2 8

LME design matrix for changes in 1 group
by amirhossein manzouri 02 Feb '23

02 Feb '23

External Email - Use Caution Hi, I have 2 sessions of data acquired in the same day for each participant before and after the drug intake. I wonder how to analyse this with LME tool. I create design matrix X in 2 columns, first all ones and second the time differences(which are the same) and wonder if I need to only run the model with one random effect like lhTh0_1RF = lme_mass_fit_EMinit(X,[1],Y,ni,lhcortex,3); And what would be the next steps to get the stats and sig.mgh Best regards, Amirhossein Manzouri

2 6

no change even after adding control points
by Rakshathi Basavaraju 01 Feb '23

01 Feb '23

External Email - Use Caution I have a scan where the aseg completely missed the bilateral frontal poles. After learning from the videos I added control points (points that were not 110) in a hope to expand the white matter surface. But after adding control points and rerunning aurorecon, still i do not see anychanges. i have attached a screenshot here. What is the error? Can this be fixed or should I discard it? [image: Screenshot 2023-01-27 at 3.55.16 PM.png] Dr. Rakshathi Basavaraju (M.D. Psychiatry) Assistant Professor (Department of Health Research-sponsored return-home research fellowship) Principal Investigator - OPCINSS Department of Psychiatry National Institute of Mental Health and Neurosciences (NIMHANS) Bengaluru, Karnataka, India-560029 Phone: +91-9480829898/+91-80-26972156/+91-80-26995320

3 2

Re: [Freesurfer] High resolution T1
by Glasser, Matt 01 Feb '23

01 Feb '23

External Email - Use Caution We showed improved data (more accurate surface placement leading to more accurate myelin maps) in prior publications (Glasser et al., 2013; 2014 Neuroimage). It makes a difference for cortical thickness too (e.g., falsely too low cortical thickness in auditory cortex in Glasser and Van Essen 2011 Journal of Neuroscience). That said, at least in our hands we have had some issues with FreeSurfer 7.X and the hires stream. I’ve been hoping that the FS folks have a chance to investigate that soon, but for now we have stuck with FS 6.0 for the HCP and HCP Pipelines. In general, it’s good to have at least two voxels of grey matter for even the thinnest cortical areas (e.g., somatosensory area 3b, visual area V1), which are about 1.6mm thick. This is where the HCP’s recommendation of at least 0.8mm isotropic T1w and T2w or FLAIR images comes from (and why we didn’t stick with the typical 1mm resolution recommendation that was prevalent before the HCP). Matt. From: <freesurfer-bounces(a)nmr.mgh.harvard.edu> on behalf of "Lee, Sang Hoon" <lsang2(a)med.miami.edu> Reply-To: Freesurfer support list <freesurfer(a)nmr.mgh.harvard.edu> Date: Monday, January 30, 2023 at 9:22 AM To: "freesurfer(a)nmr.mgh.harvard.edu" <freesurfer(a)nmr.mgh.harvard.edu> Subject: [Freesurfer] High resolution T1 External Email - Use Caution I understand that from Version 7.0, better high resolution surface reconstruction with -hires is done. My question is, if we use higher resolution T1, such as 0.8mm, will it improve the accuracy of the cortical thickness measurement? If yes, is there any reference? Thank you so much for your time! Sang H. Lee (Sr. Research Associate III) UM-Millier School of Medicine ________________________________ 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.

4 5

High resolution T1
by Lee, Sang Hoon 01 Feb '23

01 Feb '23

External Email - Use Caution I understand that from Version 7.0, better high resolution surface reconstruction with -hires is done. My question is, if we use higher resolution T1, such as 0.8mm, will it improve the accuracy of the cortical thickness measurement? If yes, is there any reference? Thank you so much for your time! Sang H. Lee (Sr. Research Associate III) UM-Millier School of Medicine

2 1

Watershedding for brainmask
by Rakshathi Basavaraju 01 Feb '23

01 Feb '23

External Email - Use Caution Dear FS people, A few of my brains were excessively skull stripped and I used watershedding to correct them by adjusting the h-values. After that I ran recon-all -autorecon-pial -subjid <subject name> I observe that the brainmask and aseg are remade but not the wm.mgz. Why is that so? Aseg is missing those parts of the brain which were initially excessively stripped (even after setting the brain mask alright and it looking perfect). I am attaching some pics here . Anyone have any solutions? Dr. Rakshathi Basavaraju (M.D. Psychiatry) Assistant Professor (Department of Health Research-sponsored return-home research fellowship) Principal Investigator - OPCINSS Department of Psychiatry National Institute of Mental Health and Neurosciences (NIMHANS) Bengaluru, Karnataka, India-560029 Phone: +91-9480829898/+91-80-26972156/+91-80-26995320

2 1

Skull stripping with external brainmask
by Boa Sorte Silva, Narlon 01 Feb '23

01 Feb '23

External Email - Use Caution Dear FreeSurfer experts, I dealing with some issues regarding skull stripping in FreeSurfer (v7.1.1). It seems like the conventional approaches (e.g., adjusting -wsthresh) doesn’t seem to help. As well, applying these preprocessing steps to the raw T1w BEFORE processing in FreeSurfer doesn’t seem to help either. I have tried: intensity truncation, N4 bias field correction, data normalization (with different combinations). Considering that we’re analyzing data from a large dataset (>500 scans * multiple timepoints), manual editing may not feasible. Therefore, to address poor skull stripping, I was wondering if it is possible to use a brain mask created outside FreeSurfer for the same subject to help with skull stripping. I saw in previous emails in the Archive that the steps below could work. Can someone confirm if this is still true and would work for v7.1.1? 1. run recon-all (as one normally would) 2. apply mri_mask to mask the T1.mgz with external brain mask volume (in native space) 3. name the output “brainmask.mgz” and move it to "subjid/mri/“ 3. re-run recon-all starting from -autorecon2 -autorecon3 Thanks for your help! Nárlon Cássio Nárlon Cássio Boa Sorte Silva PhD (He, Him, His) CIHR and MSHR BC/PARF Postdoctoral Research Fellow Aging, Mobility, and Cognitive Health Lab Djavad Mowafaghian Centre for Brain Health Department of Physical Therapy, Faculty of Medicine University of British Columbia Vancouver, Canada Twitter: @BoaNarlon

2 2

result question
by David 01 Feb '23

01 Feb '23

External Email - Use Caution Hi Freesurfer experts, I'm very sorry to bother you, but I am very confused with the following questions: My experimental design includes two discrete factors: with three levels (d,m,h); gender (f, m), and two Nvariables So I can get 6 classes: df, dm,mf, mm,hf,hm. I want to perform the difference between any of the classes leves analysis with regressing out the effect of gender and two Nvariables, but I don't know the rules for setting the contrasts for the F-test. The contrast matrix I used is: 1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 -1 0 0 0 0 0 0 0 0 0 0 0 0 My orders are as follows： mris_preproc --fsgd glmteach3.fsgd --target fsaverage --hemi lh --meas thickness --out lh_teach3.mgh mris_preproc --fsgd glmteach3.fsgd --target fsaverage --hemi rh --meas thickness --out rh_teach3.mgh mri_glmfit --y lh_teach3.mgh --fsgd glmteach3.fsgd --C glmteach3.mtx --glmdir group3.testA1_10sm.lh --fwhm 10 --surface fsaverage lh --eres-save mri_glmfit --y rh_teach3.mgh --fsgd glmteach3.fsgd --C glmteach3.mtx --glmdir group3.testA1_10sm.rh --fwhm 10 --surface fsaverage rh --eres-save mri_glmfit-sim --glmdir group3.testA1_10sm.lh --cache 3 abs --cwp .05 --2spaces mri_glmfit-sim --glmdir group3.testA1_10sm.rh --cache 3 abs --cwp .05 --2spaces We obtained five differential brain regions and extracted their cortical thickness. And then put the extracted cortical thickness into SPSS for general linear model analysis, but two of the brain regions were not statistically significant. Why? Is the contrast matrix I used wrong? Or other errors? Any help will be very appreciated. Best wishes

2 1

Salvaging brainmasks
by Rakshathi Basavaraju 01 Feb '23

01 Feb '23

External Email - Use Caution Dear FS, A few of my brainmasks were excessively stripped and I used watershedding to correct them. Many I could set right. But a few scans despite playing around with different h values if I cannot restore the cut parts of the brainmask what is to be done in that case? Should I discard the same? Thanks, Dr. Rakshathi Basavaraju (M.D. Psychiatry) Assistant Professor (Department of Health Research-sponsored return-home research fellowship) Principal Investigator - OPCINSS Department of Psychiatry National Institute of Mental Health and Neurosciences (NIMHANS) Bengaluru, Karnataka, India-560029 Phone: +91-9480829898/+91-80-26972156/+91-80-26995320

2 1

FreeSurfer load nifti error
by Siamak Sorooshyari 01 Feb '23

01 Feb '23

External Email - Use Caution Dear FreeSurfer, I am using the FieldTrip software to process MEG data, it involves my using the function ft_read_mri() in MATLAB to read in the anatomical MRI file. I have tried feeding into this function a .nii.gz as well as the unzipped .nii version. In the latter case I get an error: ERROR: D:\...\anat\sub-0249_ses-01_run-1_T1w.nii, read in 21626880 voxels, expected 10813440 I contacted the FieldTrip folks and they mentioned that it's a FreeSurfer issue. I think I've isolated the issue to the load_nifti.m file with the "nitemsread" variable and "nvoxels" variables not matching in value. I am not clear why this would happen, how can this be fixed? Best, Siamak

2 3

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer January 2023