Several things
- don't use --init-fsl. If you don't specify, then it will use mri_coreg, which will work better. You can also try --init-header
- your movable scan appears to be t2 weighted, so use --t2 instead of --t1
- look at the cost file (or send it). It is hard to say what the cost should be, but my guess from your data is that it should be below 0.5
- when you send pics, load the surfaces as that makes it really easy to tell whether the registration is good or not
- where these data sets acquired in the same session? If not, or if the subject moved a lot, then you might have to include an intermediate volume

On 11/8/2023 7:55 AM, Spapens, Jeltje (NP) wrote:

External Email - Use Caution

Hello Freesurfer Developers,

I have performed a boundary-based registration followed by a mri_vol2vol on a Neuromelanin sensitive MRI scan (partial FoV) with a T1 structural scan. I used this code:

Boundary-based registration in Freesurfer

Set the environment in the terminal.

bbregister --s MRS_203_recon_all --mov /Users/robspapens/Desktop/Thesis/Subjects/MRS_203/MRS_203_MT_tfl_stripped.mgz --reg MRS_203_NM2mp2rage.dat --T1 --init-fsl

mri_vol2vol --mov /Users/robspapens/Desktop/Thesis/Subjects/MRS_203/MRS_203_MT_tfl_stripped.mgz --targ /Users/robspapens/Desktop/Thesis/Subjects/MRS_203/MRS_203_upsamp.nii --o /Users/robspapens/Desktop/Thesis/Subjects/MRS_203/MRS_203_MT_tfl_regt1.nii.gz --reg MRS_203_NM2mp2rage.dat –cubic

I performed the analysis on two participants, but the brainstem does not overlap very nicely. I attached the screenshots of Freeview of the two participants with the T1 scan and the NM scan after the BBR. Our analysis focuses on the Substantia Nigra.

I also attached the full script that we use including the pre analysis.

Do you have a tip how to improve the BBR to reach better overlap.

Kind regards,

Jeltje Spapens
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer