Pittsburg Compound B (PIB) is a Positron Emission Tomography (PET) radio tracer used to image in-vivo, beta amyloid plaques which are one of the major histopathological hallmarks of Alzheimer’s disease (AD). Quantitative analysis of PIB retention requires normalisation of the PIB uptake values, to allow inter and intra subject comparisons. The standard uptake value ratio (SUVR) normalises the uptake values to the mean uptake value within a region containing nonspecific binding, the cerebellar grey matter in the case of PIB. SUVR calculations are normally performed manually on the PET image by neurologist which is time consuming and nonreproducible. In this study, we present an automatic technique to extract the cerebellar grey matter on PET images, and validate its accuracy for SUVR normalisation on a database of 45 subjects.