Hi Sid,
I am just looking at different skeletonization methods, which is basically what
you propose. The problem is that this is not a primary interest of my work so
I would prefer to have a ready-to-use-solution.
:)
Thanks anyway,
Martin
On Wednesday 27 January 2010 21:33:01 Siddharth Srivastava wrote:
> One way to reduce the dimension would to take a distance transform of the
> binary
> mask of the tract, and trace the maxima (someting like the medial axis
> Xform, MAT). Then,
> locally, sample within the inclusion of the mask and your FA map, along a
> plane
> perpendicular to the local direction of the medial axis. Statistics from
> these samples
> can them be derived.
> Obviously, branches in the MAT may complicate matters, but one can always
> code for exceptions.
>
> sid.
|