Hi Paul
the 130K vector is a scalar for each vertex, so there is no projection in freeview - it was already done by mri_vol2surf. Using mri_vol2surf you get to pick your projection method (check out the help). To convert from sphere to volume, you simply find the vertex on the ?h.sphere.reg file that is closest to you given (theta, phi), then look up the x,y,z position of that vertex in the ?h.white or ?h.pial surface (or somewhere between them if you want, depending on how you want to do the projection). Make sense?
cheers Bruce
On Thu, 3 Jul 2014, Paul Quigley wrote:
Hello,
Could anyone give me some insight into how freeview projects an overlay onto a surface? For example, I have an overlay generated though mri_vol2surf for subject pq, call it stats.nii. Opening the nifti in MATLAB, I see that the image it stores is a single, flat vector (had about 130,000 entries). I can overlay stats.nii onto a variety of surf files (e.g. sphere.lh, inflated.lh).
So it seems clear that each surf file has its own method for translating the indices in the vector to points on the surface... Is there any way to extract this information from the surf file?
Basically, my goal is to figure out how I can translate coordinates on sphere.?h (theta, phi or x,y,z doesn't matter) to subscripts on my T1.
Any help would be greatly appreciated!
Thank you, Paul