New Functionalities of SINDBAD Simulation Software for Spectral X-Ray Imaging Using Counting Detectors with Energy Discrimination

The computer simulation of realistic radiographs has been proved to be an efficient tool for the design of X-ray systems, the specification of their components, the development of dedicated image processing methods and the final validation. The radiographic simulation software Sindbad has been developed in our laboratory for that purpose. It allows the modeling of any radiographic set-up, any object to be examined, and takes into account the various X-ray/matter interactions. Until now, Sindbad proposes various models of detectors; but all of them provide a single value, corresponding to the deposited energy integrated over the energy range. Recently emerged semiconductor based X-ray detectors (CdTe or CZT) offer new capabilities in counting mode with energy discrimination. They can potentially decrease image noise at constant dose, enable material identification or accurate quantification. The purpose of this study is to describe recent developments of Sindbad allowing taking into account such energy sensitive detectors. These ones have been modeled, considering both physical and electronics phenomena. The integration into Sindbad offers spectral imaging simulation functionalities in both radiography and tomographic mode. Examples highlight the energetic dependency of contrast. Validation has been performed, using for comparison semi-conductor detectors developed in our laboratory. An example of quantification of beam-hardening effect illustrates the influence of detector characteristics on reconstructed image quality.