Statistical Based PET Partial Volume Correction

The partial volume effect in Positron Emission Tomography (PET) is a problem for quantitative radiotracer studies. This effect results from the limited spatial resolution of the imaging device (a few mm’s). Two factors are involved; the spillover of radioactivity into neighbouring regions and the underlying tissue inhomogeneity of the particular region. Linear methods are currently used to correct for this effect on a regional level, using tissue classification from higher resolution imaging modalities, e.g. Magnetic Resonance Imaging (MRI), which allows for the definition of anatomical regions which are assumed to be homogeneous in tracer concentration. We extend these methods to incorporate the underlying noise structure of the PET tomograph measurements, and develop fast algorithms to facilitate the computation of true tracer concentration estimates and their associated errors. We also investigate the possibility of using the noise models to infer whether the defined regions are homogenous.