Monte Carlo Simulation of a-Se X-ray Detectors for Breast Imaging: Effect of Nearest-Neighbor Recombination Algorithm on Swank Noise

摘要

We study the effect on Swank noise of different recombination algorithms for secondary carriers implemented in ARTEMIS, a detailed Monte Carlo transport that simulates the three-dimensional spatial and temporal transport of electron-hole pairs in semiconductor x-ray detectors. Two modeling approaches for recombination are compared including a first-hit (FH) algorithm that recombines the first pair from the list of candidate carriers, and a more realistic nearest-neighbor algorithm (NN). We report simulated pulse-height spectra (PHS) in a Se detector for two clinical mammography spectra, and use the entire PHS distribution to calculate Swank noise. We found that the FH and NN recombination results in terms of pulse-height spectra and Swank noise agree within the mammography energy range. The NN algorithm increased the simulation time by 30% compared to FH at 4 V/μm applied bias and 10% at 30 V/μm.