Alternating minimization algorithms for transmission tomography using energy detectors

Expectation-minimization algorithms for transmission tomography are extended to account for detectors that are modeled as integrating the energy of the incident X-ray photons. The resulting algorithm is guaranteed to monotonically increase the log likelihood functional. Both the expectation step and the maximization step are given in closed form. The formulation is general enough that it can be extended to other energy-dependent models for the detectors.