Dear Narlon,

I believe you can add -xmask [mask_volume_file] to your recon-all command.

To generate such external mask, maybe you could use mri_synthseg with the --robust flag and binarize the output?

Cheers,

/Eugenio

--

Juan Eugenio Iglesias

http://www.jeiglesias.com

From: <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of "Boa Sorte Silva, Narlon" <narlon.silva@ubc.ca>
Reply-To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Date: Monday, January 30, 2023 at 01:14
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Skull stripping with external brainmask

        External Email - Use Caution        

Hi there,

I just wanted to follow-up on this.

Thanks!

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

On Jan 18, 2023, at 7:33 PM, Boa Sorte Silva, Narlon <narlon.silva@ubc.ca> wrote:

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