External Email - Use Caution
Dear Petsurfer experts, I used partsurfer in FS6 and had applied partial volume correction (PVC) on PET data. I followed the same steps as in wiki which was very well explained! Thank you!. The analyses went fine... I would like to inquire about some issues, and I highly appreciate any clarification: 1. The flag --psf in "mri_gtmpvc": Is the value of "psf" means the amount of smoothing (fwhm in mm) that will be applied on the corrected maps? If I give "psf" a value of "zero", does this means that the corrected maps are no longer going to be smoothed? The motivation of the question is that I would like to smooth the images at the end of the analysis, so I can apply the amount of smoothing on the PVC images vs analysis involve non-pvc images.
2. For some of my group analyses, I don't see any difference between the groups. When I apply PVC using "mri_gtmpvc", I see difference in pathologically relevant regions, and vice versa for some other analyses where I see difference between the groups without PVC then the difference disappear following PVC. I highly appreciate if you please clarify why PVC can find results or eliminate findings. What is the idea of partial volume effect PVE in this case?
3. Group analysis: following PVC, I used "mri_concat" to combine the images "?.mgx.gm.nii.gz", then I smoothed the final combined images using the flag "fwhm" in "mri_surf2surf". I used fwhm=3mm and then fwhm=6mm. For all analyses I can see clusters of difference between the groups at fwhm=3mm. For some analyses the difference disappears when I use fwhm=6mm . Is this till a partial volume effect even after applying pvc or this is relevant to an extra smoothing effect which obscure the PET signal? is there any recommended fwhm value for PET images?
Thank you for your guidance, John