Skip to Main Content
The recent development of new prostate tracers has motivated us to build a low cost PET camera optimized to image the prostate. Coincidence imaging of positron emitters is achieved using a pair of curved detector banks. The bottom bank is fixed below the patient bed, and the top bank (which is above the patient) moves upward for patient access and downward for maximum sensitivity. In this paper, we study the design of septa for the prostate camera using Monte Carlo simulations. The system performance is measured by the detectability of a prostate tumor and by the conventional noise equivalent count (NEC). We have studied 27 septa configurations. The results show that septa design has a large impact on the lesion detection at a given activity concentration. At the background activity level of 0.1 μCi/cc, sparse septa with 8-crystal spacing outperforms the traditional two-dimension (inter-plane septa) and three-dimension (septaless) designs in terms of both the lesion detection and NEC. Significant differences are observed between the lesion detectability and NEC performance, indicating that the NEC may not be suitable for this lesion detection task.