[VoxBo] reorienting

[Daniel Y Kimberg]

Ling Wong wrote:

> We've done a fair amount of processing on our data (individual glms and
> rfx), but were wondering how to reorient our data into the coronal plane at
> this point.  I've tried re-orienting one subject's Anatomical into RIP, and
> then Functional, to create a new Display, but when I display the data (from
> RPI tes files), the brains appear squished from the left in right, and I'm
> not sure if Display realizes that the Display.cub and the data are in
> different orientations.  I'm just wondering, what is the easiest way to view
> the data coronally, or do I actually have to go back and reorient before
> preprocessing?

Neither the IDL-based voxel-surfer nor vbview will re-orient anything
if the orientations of your stat map and anatomical don't match.  If
you want to do serious voxel surfing in coronal orientation, your best
bet would be to re-orient your tes files and re-run your GLMs.  You
wouldn't need to re-do your preprocessing, but you would want to
create an appropriate coronal Display.cub.  By appropriate, I mean
with high in-plane resolution and the same resolution as your
functional data between planes (typically 1mm I-S and L-R, and 3mm
A-P).  You may be able to do that automatically with something like
the following procedure:

1. make a temporary directory with copies of your Anatomical.cub and Functional.cub
2. re-orient both to RIP
3. run newprep.sh

Use vbview to see if it worked, by loading functional.cub on top of
display.cub as a stat map.  Play with the threshold until you can see
if it overlays well.

If that doesn't work, you may need to play with the corner position in
the header to make sure it's correct after re-orienting.

dan