The effect of detector performance on high count rate PET imaging with a tomograph based on position-sensitive detectors

Abstract

Emission-computed tomography systems using large position-sensitive detectors must handle high event rates per detector during dynamic studies. To determine the ability of their positron-emission tomograph (PET) to perform high-count-rate scans, the authors have studied detector performance up to single rates in excess of 2 million counts per second per detector. To evaluate the quality of images resulting from data acquired at these rates, signal-to-noise ratio analyses were applied to both simulated and experimental high-count-rate images of a cold-spot phantom. Changes in image quality were quantitated, allowing the testing of combinations of high-count-rate event processing techniques to produce optimal high-count-rate images. It was found that severe image degradation occurs when the centroid positioning algorithm is used, due to the long tails in the detector point speed function that result at high count rates. Image quality is greatly improved when the local centroid event-positioning algorithm is used to eliminate the tails. Further improvements are possible with shorter integration time and an upper-level energy discriminator set at 600 keV, close to the photopeak.