Hi -

I know there is a new version of caret in the works, and I know that it is
part of the connectome project, and therefore will likely have more
features related to tractography. I'm looking forward to that hugely.

I was wondering what features it might have regarding segmentation. Caret
is the easiest, fastest and most reliable way to segment an MRI to a good
surface (certainly for macaque) that I've worked with. But the segmentation
is midthickness.

Will the new version segment at the GM/WM boundary and make midthickness
and pial surfaces from there?

When doing surface based tractography with FSL, using a WM surface that's
registered to F99 is really great. But hard to make (one must do it in
freesurfer, which is not designed for monkey).

And, I was wondering: with good quality diffusion data, the GM/WM boundary
region is often visible as an identifiable band in FA, westin and other
tensor maps.

In an ideal world, one would position the surface exactly on that band so
that the uncertainty at the boundary can be avoided when seeding/targeting
cortical regions from surface data. Perhaps there would be a way to do that?

It would be of tremendous interest to me to know what plans there are in
regard to this, and if, in general, the new caret will drop anything in
terms of features that is relevant to segmentations, surface registration
and all things macaque. And, if things are dropped whether caret5 will be
maintained.

best,

Colin
_______________________________________________
caret-users mailing list
[email protected]
http://brainvis.wustl.edu/mailman/listinfo/caret-users

Reply via email to