Hi Rob,
Maybe Daniel can verify this, but it looks fine to me. The
function that is complaining applies the Mosaic Center, which
looks like the correct thing to do (particularly since the
centers are the same, except for a clearly useless x-coordinate
in the Slice-based Center).
So I do not think there is anything to worry about, in that
regard. You might consider using 3dWarp to apply the oblique
transformation to the EPI data, just to verify that it overlays
the anatomy reasonably well.
- rick