Dear Fellow SPMers,
One of our researchers here was recently asking how to determine estimated
Brodmann's areas using the coordinates from his SPM analysis. This
prompted me to look again at the thorny issue of the differences between
the MNI brains (which SPM96 and SPM99 use as templates) and the brain in
the Talairach atlas. These differences can mean that coordinates relative
to the MNI brain, such as those given by SPM, cannot be straighforwardly
related to the Talairach atlas; This is a topic which has been discussed
fairly fully last year on the list.
I return to the topic for two reasons. Firstly, it seems to me that many
users of SPM appear to report that their data have been normalised to the
brain in the Talairach atlas. I wonder whether we the SPM users
could develop some way of pointing out to unwary readers that the coordinates
we report from SPM analyses on data normalised to the MNI brain may not be
directly comparable to the Talairach atlas.
The second reason was that I have tried to come up with a very simple little
transform that does some kind of a job of transforming coordinates from
the MNI brain to those which match the Talairach atlas, and I would be
very grateful for any feedback or comments. I have tested the transform
against the only other one that I know of, kindly posted by Andreas
Meyer-Lindenberg last year, and believe that it does a better job of matching
the two brains. The new transform, and some description of the problem,
is available on a web page:
http://www.mrc-cbu.cam.ac.uk/Imaging/mnispace.html
Once again, I would be very grateful for any comments,
Thanks a lot,
Matthew
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|