> Thanks. Can you help us in another matter.
>
> We are trying to import our SPGR structural image into SPM to try to
> coregister the fMRI map on to it. The idea is to create a mean image of
> all the structural scans form our subjects and coregister the mean image to
> the activation map. However when we try to normalize this image to MNI
> space the resultant image look quite disotorted and elongated. Could you
> let us know what is happening. Another concern is that the EPI image that
> we import and normalize may be getting disotorted too but that would be
> difficult to detect due to the poor resolution of those images.
It looks like the estimation is going wrong. This could be due to poor
starting estimates. Before trying to spatially normalise, you need to
ensure that the images are roughly aligned with the templates supplied
with SPM. Use Check Reg to do this. If you need to do any repositioning,
then use the Display button. A search through the mailing list archives
for Reorient and Display will find you plenty of hints about how to do this.
Also check that the contrasts in your image is similar to that of the
template, and that you have correct voxel sizes etc, and that the background
of the image is zero (in Display) and the forground is greater than zero.
If this is not the cause of the problem, then you could try tweeking various
other estimation parameters (e.g. brain masking [in SPM99] or Weight template
when registering [in SPM2]).
Best regards,
-John
--
Dr John Ashburner.
Functional Imaging Lab., 12 Queen Square, London WC1N 3BG, UK.
tel: +44 (0)20 78337491 or +44 (0)20 78373611 x4381
fax: +44 (0)20 78131420 http://www.fil.ion.ucl.ac.uk/~john
|