The Jacobian determinant maps are in the population average space
(rather than MNI space). You could fit a linear model to them, which
would essentially be a TBM analysis, but note that the results would
not be in MNI space.
It would be relatively easy to shift the jac_*.nii to MNI space via an
affine transform, which would just be a matter of changing the
voxel-to-world mappings in their headers. If you have run "Normalise
to MNI" on any of the data, you should have a Template_6_2mni.mat
file, which contains an appropriate affine transform.
t=load('Template_6_2mni.mat');
M=t.mni.affine;
P=spm_select(Inf,'^jac_.*\.nii');
for i=1:size(P,1),
Nii=nifti(P(i,:));
Nii.mat = M;
Nii.mat_intent = 'MNI152';
create(Nii);
end
Best regards,
-John
On 21 September 2012 12:50, alcolia <[log in to unmask]> wrote:
> Hi SPMers,
>
> I would like to perform a DBM analysis between two groups. If I am right, I realise next steps:
>
> 1. Segment my data in GM, WM and CSF.
> 2. Run DARTEL with rc* outputs from step 1.
> 3. Calcule jacobian determinant of "u_c1*".
>
> Well, I am not sure which is the next step. Should I warp "jac_*" file with the flow field or is the output from "Jacobian Determinant" warped in standard space?
>
>
> Thanks in advance
>
>
|