Although Dartel outputs are in the 1.5 mm rigid aligned space, the software keeps a track of the relative positions of these data by making use of a second affine transform matrix in the NIfTI headers of the u_ and rc files.
Best regards,
John
Paul Chou <[log in to unmask]> wrote:
>Dear John
>
>I have a question about the longitudinal VBM pipeline which you suggested in this thread.
>
>1) Longitudinal Toolbox to align each subject's data longitudinally and generate a shape average as well as Jacobain change rate maps.
>2) Segment each subject's average (default segment).
>3) Multiply the Jacobian change rate maps from (1) by the grey matter from (2) (possibly via ImCalc).
>4) Dartel to estimate how to warp all the subjects together.
>5) Dartel's normalise to MNI space using the output from (3).
>
>For the step 5, could I use the result of step3 directly? I think the result of step3 was locate in native space of each subjects and the flow field which calculated at step4 was located in the rigid-align space. Do I need to perform some additional transformation to the result of step3 into the rigid-align space or the "Normalize to MNI space" module of DARTEL toolbox could deal with this situation directly?
>
>Thank you !
>
>Best
>
>Paul
>
|