Automatic determination algorithm of intrinsic parameters on response function of energy-resolving photon counting imaging detector

In medical X-ray diagnosis, the Energy-Resolving Photon Counting Detector (ERPCD) technology has focused on deriving quantitative information of objects. For the quantitative analysis using polychromatic X-rays, the effects of beam hardening and the detector response should be corrected. To estimate the response function accurately, it is necessary to determine the optimal combination of the two parameters related to energy resolution (peak broadening) and charge-sharing effect (peak efficiency), which are the intrinsic characteristics of each detector module. However, to manually tune the parameters of modules takes a lot of time and therefore it is impractical. This study aims to develop an automatic determination algorithm for intrinsic parameters. We focus on the fact that the measured X-ray attenuation rates (µt values) of objects reflect the parameters of the detector module. In our algorithm, the proper parameters can be determined so as to reproduce the experimental µt values among the simulated µt values considered various parameters. Experimental values were measured using an aluminum step phantom with 1–30 mm thicknesses and our ERPCD system, which consists of CZT multi-pixel detector with a pixel size of 200 µm × 200 µm and three energy bins. For each parameter, we evaluated the differences between the experimental and theoretical µt values for three energy bins: 20–32–40–56 keV for 60 kV radiographic system. The algorithm was verified by deriving the effective atomic number (Z
eff
) image of the aluminum step phantom and evaluating the accuracy. As a result, the accuracy of Z
eff
values was approximately ±1.0, suggesting that our algorithm can be used to tune an imaging detector for quantitative analysis in radiography. The developed algorithm is expected to make it possible to quickly and automatically determine the response function parameters unique to each module.