[caret-users] white matter point on flattened surface

[Donna Dierker]

Hi Sofia,

Let's say you have an activation map (fMRI) in volumetric form for your 
subject, for whom you also have a fiducial surface.  Provided that the 
fMRI has been volume-registered to the same anatomical used to generate 
the fiducial surface.

When you map that fMRI volume onto the surface, you always map to the 
fiducial surface.  (Note that the fiducial surface might be a white 
matter, pial, or midthickness representation.  If the surface was 
generated using Caret, it's a midthickness surface.  If it was generated 
using, say, Freesurfer, then a midthickness surface is easily generated 
by averaging the WM and pial surfaces.)

Once the fMRI is mapped onto the fiducial surface, then you have a 
node-scalar mapping that can be rendered/overlaid on any surface 
configuration (flat, inflated, etc.).  But one never maps to the 
inflated/flat surfaces directly.

It's worth mentioning that the Caret surface, being a midthickness 
surface, won't necessarily intersect your white matter voxel.  It 
depends on the fMRI resolution and how the fiducial surface intersects 
the fMRI volume there.  There are various mapping algorithms that help 
you get around this problem (gaussian being the most helpful, 
probably).  But if you have an actual white matter surface available, it 
is possible to map to that, as well.  If the voxel is deep in the white 
matter, it will still not intersect the WM surface, but then it is of 
questionable validity, I would think.

Hope this helps,

Donna

On 05/14/2009 04:45 AM, s.crespi1 at campus.unimib.it wrote:
> dear Donna
> I have a we question concerning the overlapping metrical data on the flat map.
> If I have some voxel belonging to the white matter how does caret manage them with respect to the flat?
>
> thank you!
> sofia
> _______________________________________________
> caret-users mailing list
> caret-users at brainvis.wustl.edu
> http://brainvis.wustl.edu/mailman/listinfo/caret-users
>