Hi Chris,
The value in seeds_to_blah is the number of samples reaching an ROI.
So it doesn't matter if the ROI is hollow or not (unless the step size
is bigger than a voxel).
Saad.
On 20 Dec 2009, at 11:49, Chris Lambert wrote:
> Seasons Greetings FSL'ers
>
> I am interested in how the topology of a classification target could
> influence my results. FDT classification produces a single voxel
> value for
> each one in the mask, but depending on the size of the
> classification target
> many of the seeds from a single voxel could reach and pass through
> it (for
> example if one looked at thalamic mask vs. putamen (3D ROI)
> classification).
> Does the voxel value in seeds_to_blah represent the number of seeds
> reaching
> the edge of the ROI (so the outer shell of ROI voxels) or is it the
> sum of
> connectivity values within the entire classification target for a
> single
> voxel (i.e. Would I get different results with surface "shell" ROIs
> vs.
> dense (whole volume) ROIs)?
>
> Many thanks,
>
> -Chris
>
--
Saad Jbabdi
University of Oxford, FMRIB Centre
JR Hospital, Headington, OX3 9DU, UK
+44 (0) 1865 222466 (fax 717)
www.fmrib.ox.ac.uk/~saad
|