Combination of high resolution analytically computed uncollided flux images with low resolution Monte Carlo computed scattered flux images

A new computing model has been implemented in Sindbad, a simulation software dedicated to radiographic systems, to provide quite rapidly realistic radiographs of complex objects, including the scatter component of the photon flux. The new computing model consists of a coupling of two previous models: an analytical approach and a Monte Carlo one. The analytical simulation is used to compute a high resolution uncollided photon flux image whereas the Monte Carlo provides the scattered flux image. This latter image is computed as a scaling of a reduced fluence Monte Carlo scattered flux image. Scaling is possible thanks to the low resolution feature of the scattered flux image. After a detailed description of the method developed to perform the combination, we present a validation on a medical radiograph simulation. In spite of the quite complex geometry of the examined part and the large number of emitted photons, this new computing model provides a realistic radiograph with a good estimation of the scattered flux in a reasonable computation time.