To warp findings back to an individual subject, you need the inverse
of the mapping originally used to spatially normalise the data. In
principle, you could do this via the "Normalise: Write" option
(specifying the T1w_seg_inv_sn.mat), although this may not quite give
what you want if the original data are not exactly axial.
Instead, I'd suggest going via the Batch system, selecting
SPM->Utils->Deformations.
Composition
Imported _sn.mat
Parameter File T1w_seg_inv_sn.mat
Voxel sizes [NaN NaN NaN]
Bounding box [NaN NaN NaN; NaN NaN NaN]
Identity (Reference Image)
Image to base Id on the EPI image you want to align to
Save as
Apply to mask.img
Best regards,
-John
On 4 September 2012 12:21, Karin Lundengård <[log in to unmask]> wrote:
> Hi!
> I have a series of fMRI images which I have used SPM8 to realign, coregister
> and smooth and from there I have made a GLM group analysis. From the
> smoothed group image I have now used MarsBar to make masks which I have
> transformed to .nii so I can get them as matrices in matlab. Now I want to
> use the transform matrices from the coregistration to transfrom these masks
> back to the same format as the realigned images. How do I do that? I am
> guessing that I should use the files from the coregistration T1w_seg_sn.mat
> and T1w_seg_inv_sn.mat (my anatomical image is named T1), but what do I do
> with them?
> Does anyone know?
> /Karin
|