A one-step algorithm for spectral CT with an application on multi-source inverse geometry

International audienceSpectral computerized tomography (Spectral CT) is an imaging technique which uses the spectral information of the attenuated X-ray beam. Energy-resolved photon-counting detector is a promising technique for improved spectral CT imaging and allows to obtain material selective images. While energy-resolved photon-counting detectors can have good spectral resolution (with many energy bins), the size of these detectors is often limited. For their part Multi-Source Inverse Geometry CT (MS-IGCT) architectures allow the use of a smaller detector than conventional Cone Beam CT architecture (CBCT). In a previous work, we have proposed a one-step reconstruction algorithm validated with spectral data from classical Cone Beam CT architecture. In this work we propose to adapt this one-step method on spectral data from MS-IGCT architecture. From noisy simulated data, we compare the proposed one-step method with two two-step decomposition methods