We consider penalized-likelihood reconstruction for X-ray computed tomography of objects that contain small metal structures. To reduce the beam hardening artefacts induced by these structures, we derive the reconstruction algorithm from a projection model that takes into account the photon emission spectrum and nonlinear variation of attenuation to photon energy. This algorithm requires excessively long runtimes to attain convergence. We contribute refinements in the representation of the object and the modelling of the photon emission spectrum aimed at reducing the runtime. These refinements yield an algorithm for which the reconstruction runtime is reduced by at least one order of magnitude on each considered dataset. For real data that were preprocessed for metal artefact reduction, the enhanced polychromatic algorithm provides the best observed image quality for objects with large metal structures. For objects with small metal structures, the gains are less significant with respect to a penalized-likelihood algorithm derived from a monochromatic projection model.
École Polytechnique de Montréal, December, 2009
View Metal Artefact Reduction by Least-Squares Penalized-Likelihood Reconstruction with a Fast Polychromatic Projection Model