A NONLINEAR SPATIALLY VARIANT OBJECT-DEPENDENT SYSTEM MODEL FOR PREDICTION OF PARTIAL VOLUME EFFECTS AND SCATTER IN PET

Accurate quantitation of small lesions with positron emission tomograp hy (PET) requires correction for the partial volume effect. Traditiona l methods that use Gaussian models of the PET system were found to be insufficient. A new approach that models the non-Gaussian object-depen dent scatter was developed. The model consists of eight simple functio ns with a total of 24 parameters. Images of line and disk sources in c ircular and elliptical cylinders, and an anthropomorphic chest phantom were used to determine the parameter values. Empirical rules to deter mine these parameter values for various objects based on those for a r eference object, a 21.5-cm circular cylinder, were also proposed. For seven spheroids and a 3.4-cm cylinder, pixel values predicted by the m odel were compared with the measured values. The model-to-measurement- ratio was 1.03 +/- 0.07 near the center of the spheroids and 0.99 +/- 0.03 near the center of the 3.4-cm cylinder. In comparison, the standa rd single Gaussian model had corresponding ratios of 1.27 +/- 0.09 and 1.24 +/- 0.03, respectively, and the corresponding ratios for a doubl e Gaussian model were 1.13 +/- 0.09 and 1.05 +/- 0.01. Scatter fractio n (28.5%) for a line source in the 21.5-cm cylinder was correctly esti mated by our model. Because of scatter, we found that errors in the me asurement of activity in spheroids with diameters from 0.6 to 3.4 cm w ere more significant than previously appreciated.