Pinhole edge penetration and scatter in small-animal energy-integrating pinhole emission computed tomography

In small animal emission computed tomography, multiple-camera systems can provide high sensitivity and rapid imaging. With enough cameras, all views can be acquired simultaneously, with no camera rotation. Such a system operated in SPECT mode is expensive because of the cost of the energy-discriminating cameras required for scatter rejection. The scatter from mice and rats is far less than that from humans, reducing the need for scatter rejection. A small-animal ECT system based on energy-integrating detectors could be much more affordable and widely available than a multi-camera SPECT system. Previous Monte Carlo studies have shown that an energy-integrating detector, accepting all scatter, would have a 20% lower effective photon count (based on SNR) than a detector with perfect energy discrimination when imaging a 2-cm diameter water cylinder with Tc-99m. The reduction would be as much as 60% for a 5-cm diameter cylinder imaged with I-125. Signal errors due to scatter would be comparable and would present a challenge for reconstruction algorithms. The effects of scatter and penetration at pinhole edges are well recognized, but have generally been studied with the assumption that scattered photons could be rejected by energy discrimination. This study examines the resolution and noise effects of pinhole edge penetration and scatter for tungsten, gold and platinum pinholes for energy-integrating systems, with effects for energy-discriminating systems examined for comparison.