Simulation of the effects of multiplexing in multi-pinhole SPECT using stacked Si-HPGe detectors

Small-animal pinhole single photon emission computed tomography (SPECT) image quality is limited by a number of factors, including the system's detection efficiency. Efficiency can be improved upon by using multi-pinhole apertures, but when detector space is limited pinhole projections may overlap, or multiplex, which can cause artifacts in reconstruction. We are currently developing a SPECT system utilizing a high-purity germanium (HPGe) detector, but due to its limited size, in order to achieve reasonable magnification with multiple pinholes, projections on the HPGe are likely to have some amount of multiplexing. In this work we explore whether a stacked-detector configuration with both HPGe and a silicon detector, used with 1231 (27-32, 159 keY), where little or no multiplexing occurs in the Si projections, can help to compensate for the image degradation caused by the multiplexed HPGe projections. Simulations with varying pinhole configurations in conjunction with three different object phantoms (a hot-rod, cold-rod and cool-spot) are used to examine whether the additional projections from the Si detector help to overcome the artifacts from the multiplexing on the HPGe detector. The number of pinholes and the distance between pinholes are varied such that different amounts of multiplexing are seen on the HPGe detector. Reconstructed images using both Si and HPGe data are compared to those using HPGe data alone. Normalized mean square error provides a quantitative comparison of reconstructed images to the original phantoms and a means to evaluate the impact of the additional non-multiplexed data on image quality. For a qualitative comparison, the differential point response function is used to examine artifacts for each configuration. Results show that in cases of highly-multiplexed HPGe projections, the addition o