Neither reason really holds up. The surface vertex does not have a volume associated with it. The extra vertices just look like smoothness in the correction for multiple comparisons (which you will have regardess), so there is no penalty. Bottom line is that you should stick to fsaverage ico7
On 2/17/2020 5:35 AM, Marina Fernández wrote:
External Email - Use Caution
Hi Doug,
We thought it would be better to use 5th order icosahedron tesselation for PET because the resolution of the volume is lower: If, for thickness analysis, we carry the structural volumes with a resolution of 1x1x1 to ico7 on the surface, with PET-volumes or connectivity maps (resolution of 2x2x2 or 3x3x3) we thought it would be more appropriate to use ico5 on the surface because in the surface transformation the interpolation of the data is lower. In addition the number of statistical comparisons is greatly reduced.
Regards, Marina.
Marina.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Hi experts, This is a question related to PetSurfer steps.
I would like to know if it is correct to sample the mgx volume onto the surface of the average subject of my dataset (created with 5th order icosahedron tesselation) instead of the fsaverage (created with 7th order icosahedron tesselation) or if there is any problem in the next steps following this procedure. We would like to do that because the volume PET resolution is more proportional to this number of vertex (10242). Do you think it is correct?
Best wishes, Marina.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
You could do it, but I would just use fsaverage. I don't understand what the relationship would be between the number of vertices on the surface and the number of voxels in the PET volume or the rational for using the 5th order ico
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer