In-silico optimisation of tileable philips digital SiPM based thin monolithic scintillator detectors for SPECT applications

Highlights: • Geant4 was used to optimise SiPM based thin monolithic scintillator SPECT detectors. • For Philips SiPMs, four materials and impact of their crystal thickness was explored. • A critical minimum crystal thickness of 3 mm was determine for all materials. • CsI(Tl) was determined to represent the most promising material for Philips SiPMs. Over the last decade one of the most significant technological advances made in the field of radiation detectors for nuclear medicine was the development of Silicon Photomultipler (SiPM) sensors. At present only a small number of SiPM based radiation detectors for Single Photon Emission Computed Tomography (SPECT) applications have been explored, and even fewer experimental prototypes developed. An in-silico investigation into the optimal design of a Philips DPC3200 SiPM photosensor-based thin monolithic scintillator detector for SPECT applications was undertaken using the Monte Carlo radiation transport modelling toolkit Geant4 version 10.5. The performance of the 20 different SPECT radiation detector configurations, 4 scintillator materials (NaI(Tl), GAGG(Ce), CsI(Tl) and LaBr₃(Ce)) and 5 thicknesses (1–5 mm), were determined through the use of seven figures of merit. It was found that a crystal thickness range of 4–5 mm was required for all four materials to ensure acceptable energy resolution, sensitivity and spatial resolution performance with the Philips DPC3200 SiPM. Any thinner than this and the performance of all four materials was found to degrade rapidly due to a high probability of material specific fluorescence x-ray escape after incident gamma/x-ray photoelectric absorption. When factoring in each material's magnetic resonance imaging compatibility, hygroscopy, and cost, it was found that CsI(Tl) represents the most promising material to construct tileable Philips digital SiPM based thin monolithic scintillator detectors for SPECT applications.