I am quite inexperienced with the technology but have been asked to use PETSurfer to create a 'SUVR Image' of an FDG Pet scan, where the number associated with intensity in the motion-corrected, partial volume corrected value is the SUVR value. 

>From what I understand, an indication of success would be that the values on the mgx.gm.nii.gz being on the scale of around 1-10, rather than in the thousands. I have tried to achieve this by adding the --rescale 7 8 46 47 flag to the partial volume correction step, but the numbers are still in the order of the thousands. Is there any way to manually rescale the file's intensities, dividing the entire file's intensities by the average value across the cerebellum?

Thank you for your help, please let me know if my question is unclear.

Ian