I’m not taking sides on this one, but I’ll nevertheless pinpoint that you can also get a sbTIV using SynthSeg, hehe. Looking forward to reading people’s opinions on this one!
-- Juan Eugenio Iglesias http://www.jeiglesias.com
From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Jackson Lee jacksonml@student.unimelb.edu.au Date: Tuesday, January 30, 2024 at 7:29 PM To: freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Choice of ICV estimation
External Email - Use Caution Hi there,
My question is related to FreeSurfer’s calculations for intracranial volume (ICV): Given the importance to consider ICV as a potential covariate when assessing volumetric differences in ROIs between groups, which ICV output is recommended?
FreeSurfer’s recon-all generates eTIV (which is widely used by default in literature, and is a registration-based method). Conversely, the SAMSEG pipeline generates sbTIV (relatively newer approach which is a segmentation-based method). While these two would be the most obvious choice, many papers also consider correcting for supratentorial volume to be an appropriate approach.
To some degree, the choice of method would be guided by the research question. However in the most simple of cases (e.g., non-clinical cohorts), what method is recommended? Why would someone, for example, opt to use sbTIV over eTIV?
Warm regards, Jackson Lee