Hello there, (and Hello rsmith)
I am interested in mesh2voxel in order to quantify partial volume of gray matter (GM)
(related post is Partial Volume and (linear) interpolation · nipy/nibabel · Discussion #1264 · GitHub)
I guess this is a borderline subject, as it has not too much to do with diffusion, but mesh2voxel, is a nice feature, and I could not find many equivalent elsewhere
There is one equivalent in freesurfer, although it only produce a binary file (not quite sure, I need to further test, but it is quite long to run)
Other option I found is the Toblerone software, which is quite long to run too.
in comparison, mesh2voxel is super quick !
When comparing to Toblerone I noticed some errors of mesh2voxel, especially in voxels where two separate GM vertex are close (in the sulci)
I find less difference if I use mesh2voxel at a much smaller target resolution and than upsample with a factor multiple of 2 (in order to get exact partial volume)
does it make sense ?
I am just curious if you know where the errors comes from and if there may be a fix ?
Also for my understanding, why is it so difficult to compute ? (Cf toblerone article)
The fsleyes viewer can draw (very quick thanks to a python library) the surface intersection line with the slice, or the filled mask whitin the surface
if we do it for all slices, then we should get the same result ?
(may be it is an alternative to have it done quick in python ?)
Romain