Dear Nima,

Thanks for the interest in our work.

I don’t know if I would advise trying to segment these small subregions from such thick slices. But if you are set on doing that, I would probably run SynthSR first, and work with the synthetic 1mm isotropic T1s. You can run SynthSR on the T1s and T2s and keep the output that works best.

Cheers,

/Eugenio

--

Juan Eugenio Iglesias

http://www.jeiglesias.com

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Taghizadeh Mortezaei, Nima <ntaghizadehmortezaei@mgh.harvard.edu>
Date: Friday, April 4, 2025 at 1:52 PM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] Guidance on SegmentHA with Low-Res MRI

Dear FreeSurfer Community,

I’m currently working with clinical MRI data and would appreciate your guidance regarding two issues I'm encountering with hippocampal/amygdala segmentation (segmentHA_T1.sh script). My available T1w and T2w scans both have 6 mm slice thickness and are anisotropic and I/m working with FreeSurfer 7.4.1.

Given the slice thickness and anisotropy, is it advisable to use recon-all-clinical instead of the standard recon-all as a preprocessing step? I understand that the segmentation scripts depend on the subject having a full recon-all directory, but I’m unsure whether recon-all-clinical can serve as an alternative for this purpose or if the standard recon-all is strictly required for compatibility with the hippocampal segmentation pipeline.

Would including the T2 scan be helpful if it has the same 6 mm slice thickness and anisotropic resolution as the T1?

Thank you very much for your time and help.

Best,

Nima