Improving basis material decomposition in the presence of x-ray scatter with an energy-resolved photon counting detector

New horizons in x-ray imaging have been opened up with the emergence of energy-resolved photon counting detectors (PCDs). These include the ability to differentiate material components and estimate their equivalent thickness or relative ratio by processing a single shot acquisition image. However, such techniques require highly accurate images, especially for materials close in terms of attenuation. The presence of scattered radiation leads to a loss of contrast and, more importantly, a bias in radiographic material imaging and artefacts in cone-beam CT. The aim of the present study was to evaluate the performance gain in material imaging when a Partial Attenuation Spectral Scatter Separation Approach (PASSSA) is used. This evaluation was based on numerical simulations using Sindbad-SFFD. The comparison of water and bone thickness images acquired from scatter free, scatter corrupted and PASSSA-corrected images showed that the application of the considered correction method improved thickness estimation accuracy. The average relative thickness estimation error per pixel was reduced by around a factor of 4.7 for both water and bone images when comparing scatter corrupted and PASSSA-corrected cases.