Hi Carly,
SynthStrip has been included since FreeSurfer 7.3 but not as the default skull-stripping method before the 8.0 beta release. You can explicitly enable it by adding the `-synthstrip` flag to your recon-all command and setting the environment variable FS_ALLOW_DEEP to 1.
Details can be found in the release notes: https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes
Best wishes, Malte
________________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Carly McIntyre-Wood mcintc10@mcmaster.ca Sent: Wednesday, February 26, 2025 12:25 To: freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Issues with FS 7.3 skull stripping but NOT mri_synthstrip
External Email - Use Caution Hi there,
I have several participants that, when ran through Freesurfer 7.3, end of losing portions of brain matter during skull-stripping. Interestingly, when I run them through mri_synthstrip alone, this issue does not happen. If mri_synthstrip is the default for 7.3, how can this be happening? Is there perhaps a step before in the standard recon-all pipeline that is causing the issue? I’ve attached two images: one skull stripped with synthstrip and the other the brainmask.mgz file generated through recon-all.
[cid:image003.png@01DB8849.269D3A30] [cid:image002.png@01DB8849.2F975A80]
Thanks so much, Carly
FREESURFER_HOME /Users/mackilloplab/Applications/freesurfer7.3 FSFAST_HOME /Users/mackilloplab/Applications/freesurfer7.3/fsfast FSF_OUTPUT_FORMAT nii.gz SUBJECTS_DIR /Applications/freesurferOLDdownload/subjects_OLDFSDOWNLOAD/NEUROALC_SS MNI_DIR /Users/mackilloplab/Applications/freesurfer7.3/mni
Carly McIntyre-Wood, MSc Research Technician – Neuroinformatics Lead Peter Boris Centre for Addictions Research @PBCARMacSJHH St. Joseph Healthcare Hamilton & McMaster University [Peter Boris Centre (@PBCARMacSJHH) / Twitter]
7.3 does not use synth strip. That is not default until 8.0. You can try running recon-all with -xmask synthstrip.mgz and this should override the internal brainmasking.
On 2/26/2025 12:25 PM, Carly McIntyre-Wood wrote:
External Email - Use Caution
Hi there,
I have several participants that, when ran through Freesurfer 7.3, end of losing portions of brain matter during skull-stripping. Interestingly, when I run them through mri_synthstrip alone, this issue does not happen. If mri_synthstrip is the default for 7.3, how can this be happening? Is there perhaps a step before in the standard recon-all pipeline that is causing the issue? I’ve attached two images: one skull stripped with synthstrip and the other the brainmask.mgz file generated through recon-all.
Thanks so much,
Carly
FREESURFER_HOME /Users/mackilloplab/Applications/freesurfer7.3
FSFAST_HOME /Users/mackilloplab/Applications/freesurfer7.3/fsfast
FSF_OUTPUT_FORMAT nii.gz
SUBJECTS_DIR /Applications/freesurferOLDdownload/subjects_OLDFSDOWNLOAD/NEUROALC_SS
MNI_DIR /Users/mackilloplab/Applications/freesurfer7.3/mni
*Carly McIntyre-Wood, MSc Research Technician – Neuroinformatics Lead* *Peter Boris Centre for Addictions Research @PBCARMacSJHH*
St. Joseph Healthcare Hamilton & McMaster University
Peter Boris Centre (@PBCARMacSJHH) / Twitter
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu