Dear Matt,
The answer is: yes, you can coregister images with different
resolutions. This is very common with human data set including low res
functional images (EPI) and high res structural images (typically T1),
like you.
SPM accounts for the "voxel world"-to-"real world" transform, and you do
not need to resample any image before hand, neither afterwards. The
estimated rigid body transform will simply be included in the "voxel
world"-to-"real world" transform of one image, depending on which one
you selected as source and target.
Best,
Chris
====================================================
Christophe Phillips, Ir. Ph.D.
FNRS-FRS Research Associate & Assistant Professor
Cyclotron Research Centre, B30, University of Liège
4000 Liège, Belgium
tel: +32 4 366 2316/2366
fax: +32 4366 2946
email: [log in to unmask]
====================================================
R. Matthew Hutchison a écrit :
> Hi SPMers,
>
> I was wondering if SPM5 is able to coregister a functional image that
> is 64 x 64 x 13 slices to an anatomical that is 256 x 256 x 13 slices.
> The main problem being the discrepancy is the size of the images. Does
> this issue arise when using human data?
>
> When running coregistration it seems SPM5 trys to move the small image
> around and warp it within the larger space but does not interpolate it
> to the larger image.
>
> It works fine if I interpolate the anatomical to the 64 X 64 size
> before hand but I was hoping to preserve the resolution of the larger
> anatomical.
> Perhaps this is not possible though.
>
> Thanks for your thoughts,
>
> Matt Hutchison
>
> --
> R. Matthew Hutchison, PhD. Candidate
> Centre for Functional and Metabolic Mapping
> Robarts Research Institute
> Cuddy Wing - 9.4T Suite
> P.O. Box 5015, 100 Perth Drive
> London, Ontario, Canada N6A 5K8
|