Feasibility study for photon counting detector for high resolution pre clinical SPECT

We plan to integrate a photon counting, large area, high spatial resolution, high efficiency detector in a system for pre-clinical SPECT. We are aiming for a high spatial resolution for the detector of around 20 μm. The agent for radiolabelling is assumed to be ¹²⁵I, with an emission peak at 27 keV, since it is readily available. The aim of this feasibility study was to optimize the scintillator and electronics. We consider a complementary metal-oxide-semiconductor (CMOS) circuit together with a columnar CsI scintillator. In our case CMOS is advantageous over CCD due to a more flexible readout and an opportunity for data processing and trigger logic at the pixel level. The position of the x-ray interaction in the scintillator can be estimated with high accuracy by fitting a two-dimensional Gaussian curve to the pixel hit-map. The goal of this study is to determine optimal scintillator thickness, pixel size, readout rate and the corresponding achievable spatial resolution. Results for a CsI thickness of 600 μm show a detection efficiency close to 90 % and a spatial resolution in the range of 20 μm. Optimum pixel size for this application is between 40 – 60 μm and as long as the readout speed exceeds 750 Hz the noise from the dark current will not be significant.