[Freesurfer] Petsurfer: PET -> MNI305 problem

Hi Doug and experts,

We're interested in running a voxel wise test-retest reliability analysis on a set of PBR28 PET images that we have.

Following the petsurfer wiki page (https://surfer.nmr.mgh.harvard.edu/fswiki/PetSurfer), I ran this command to get one of the images into 2mm MNI305 space:

mri_vol2vol --mov mgx.gm.nii.gz --reg aux/bbpet2anat.lta --tal --talres 2 --o gm.mni305.2mm.sm00.nii.gz

However, the resulting image has 2 problems (see attached):

1) it is cutoff at the edges 2) the axes are switched, suggesting there's something wrong with the voxel ordering.

I tested the correctness of aux/bbpet2anat.lta by applying it to mgx.gm.nii.gz and it gives the correct result, i.e., it is aligned with orig.mgz.

Any ideas on what may be going wrong or suggestions on how to fix this?

Thanks! Ryan