PET system calibrations and corrections for quantitative and spatially accurate images

Detector efficiency normalization and spatial distortion corrections are considered for PET (positron emission tomography) systems. A technique is developed for correcting positioning errors. To correct for nonuniform detector efficiencies, it is necessary to normalize individual detector efficiencies with a PET equivalent of a flood field correction. The frequency distributions of the percent deviations of the processed low-count normalizations and cylinder normalizations indicate that the assumptions of the calculations entail errors that mimic 3-5% standard deviations in the final values of the normalization factors. In normal operation of the UCLA PET center, a full high-count normalization is performed, and the normalization sinogram uses the measured values in the central portion to and extends the FOV (field of view) by the method described.>