Dear Amir,

1. By default, FS will use norm.mgz, aseg.mgz, etc. You will have to eg rename norm.mgz to norm.1mm.mgz and copy / symlink / rename the _hires equivalents to get FS to use the highres T1s.
2. That sounds appropriate!
3. That sounds about right 😉

Cheers,

/Eugenio

Juan Eugenio Iglesias

Senior research fellow

CMIC (UCL), MGH (HMS) and CSAIL (MIT)

http://www.jeiglesias.com

From: AmirHussein Abdolalizadeh <amirhussein.a@gmail.com>
Date: Sunday, October 18, 2020 at 02:48
To: "freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>
Cc: "Iglesias Gonzalez, Eugenio" <e.iglesias@ucl.ac.uk>
Subject: segmentHA on Human Connectome Project

Hi,

I am trying to run segmentHA pipeline on HCP data. To do so, I have downloaded FS processed-extended structural data, manually made "scripts", "trash", and "stats" folders in each subject's folder [HCP output does not have these folders]. A few notes and questions:

1. segmentHA_T1 runs smoothly. However, I do not know which T1 does segmentHA use, since there are also a few "_hires" (high resolution) files resulting from HCP's structural analysis pipeline (https://github.com/Washington-University/HCPpipelines).

2. segmentHA_T2 also runs smoothly. However, I didn't know which T2 file is the most suitable to be given to this command. I did give T2w_hires.nii.gz (Voxelsize = 0.7 mm isotropic) in the /mri folder of each subject.

3. I statistically checked the correlation between HCP FS generated values of whole hippocampus/amygdala, and segmentHA generated ones. Of note, they were highly correlated (all p-values < 0.001, all r ~ 0.9).

I will be glad to know whether the pipeline I used to analyze HCP data is correct or any further steps are required.

Bests,

Amir