MEASURING PET SCANNER SENSITIVITY - RELATING COUNTRATES TO IMAGE SIGNAL-TO-NOISE RATIOS USING NOISE EQUIVALENT COUNTS

Sensitivity parameters derived from a plot of a scanner's true coincidence count (TCC) rates as a function of activity in a 20cm cylindrical phantom have no direct link to image quality. Noise equivalent count (NEC) rate curves, which incorporate the noise effects of subtracting the randoms and scatter count components provide a direct link between image signal-to-noise ratios and the scatter, randoms and trues coincidence count rates. We measured TCC and NEC curves with a standardized 20cm diameter nylon cylinder for five different CTI/Siemens PET scanners with several scanner-collimator combinations: (1) 831/08-12 with 1mm collimator septa; (2) 933/08-12 and 933/08-16 with 3→1mm tapered collimator septa; (3) 931/08-12 with 3→1mm tapered and a 1mm collimator septa and the 931/08-16 with 3→1mm tapered collimator septa. In addition, we compared TCC and NEC curves on the 933/08-12 with those from an Alderson brain phantom. In general we found that the TCC curves indicated peak count rates and activity levels that were as much as 50% higher than the corresponding values from NEC curves. The primary factor causing this difference is the noise effect of the randoms component. In addition, we demonstrated that compared to the Alderson brain phantom the standard 20cm cylinder is a poor predictor of count rate performance for PET brain imaging. Copyright © 1990 by The Institute of Electrical and Electronics Engineers, Inc.