Dear whomever it may concern,

I'm Chungmin, a graduate research assistant in University of Texas at Austin, I'm fairly new to using freesurfer, I'd appreciate to get some advice on what I'm facing,

*I've had this posted but for some reason lost track of it so I'm uploading it again.*

I have fMRI limited FOV collected on brainstem + little cerebellum/visual area, I've tried SPM manual registration to MNI space, falls in 1-2mm (1-2 voxels) away from accurate registering brainstem correctly, I have T1 mprage in 1mm resolution, but seems T1 registration to MNI space has a bit of error around "Brainstem" area.

is there a way to create transformation matrix *optimized to fit brainstem in MNI space*? Registering cortical region slightly bit wrong is okay to me at this moment,

https://surfer.nmr.mgh.harvard.edu/fswiki/BrainstemSubstructures

I've noticed this has been posted pretty recently, but would like to figure out how to align in standard space for comparison.

I've also thought of inverse registering from MNI to each native space, however, if the transformation focuses on maximizing registration on the cortical level (T1) I don't think inverse registration solves the issue,

The area that I'm interested in is lateral nuclei, reticular formation etc. so it would rather be focused on the midbrain, not even at the level of diencephalon,

and Additional question, would using fsaverage be more beneficial over MNI152 registration? in terms of brainstem warped into the standard space more accurately.

Thank you in advance :) Best

PS How do you reply to your own question after someone replies?

Chungmin

Chungmin Han

Graduate Research Assistant The University of Texas at Austin 204 E. Dean Keeton St Austin TX 78712 Mobile: 512-717-1671