Hi, I am running fMRIPREP for my research with preexisting freesurfer directory.
What I recently learned is that at least in the QC report at the end (html) for segmentation visualization, fMRIPREP is using FSL fast procedure.
I was searching for any method to switch this segmentation output to aseg.mgz from freesurfer output, since I am using 7t fMRI data and found out that freesurfer segmentation quality is way better than FSL fast procedure for segmentation, especially gray matter area with low signal.
While searching, I found an old post in github (sMRIPREP, 2020 that authors actually tried using aseg from freesurfer than fsl FAST when freesurfer directory is provided.
But when I checked the most recent code from sMRIPREP github, this change is not applied currently. Is this correct, and if it is, what would be the reason authors changed it back? And also, in that case, would it be okay to use that previous version of code, or at least, manually created segmentation file (using aseg) as ‘precomputed’ derivatives?