Hi -
Probtrackx apply the masking in diffusion space, but binning in seed space (in your case MNI).
So the fact that fdt_paths appear to go outside your MNI-space mask is a combination of slight mis-registration of the brain edges and interpolation due to resampling.
If you want probtrackx to stop at the edge of the MNI mask instead of the native diffusion space brain mask, you can transform the MNI mask to diffusion space (using nearest neighbour interpolation) and mask your nodif_brain_mask against it.
Cheers,
Saad.
On 23 Aug 2011, at 05:07, Walden, Eric wrote:
> When I run ProbtractX, using a mask in MNI space, the fdt_paths extend just a little beyond the brian. It's not much. I have specified that the seed space is not diffusion and used the standard2diff.mat matrix to transform it. I created this matrix using the FDT registration program with the MNI152_T1_2mm_brain and a Betted version of the diffusion image.
>
--
Saad Jbabdi
University of Oxford, FMRIB Centre
JR Hospital, Headington, OX3 9DU, UK
(+44)1865-222466 (fax 717)
www.fmrib.ox.ac.uk/~saad
|