Unified approach to regularized maximum likelihood estimation in computed tomography

Since 1982, when it was first proposed by Shepp and Vardi,(1) the Expectation Maximization (EM) algorithm has become very popular among researchers in image reconstruction. Recently, a natural extension of the EM algorithm was proposed(2) in order to handle regularization terms containing 'a priori' information far emission computed tomography (ECT) problems. This new idea was further applied to other regularized maximum likelihood problems(3-5) in transmision and emission tomography. We present in this article a unified approach to more general regularized ML problems. Our convergence proofs also extend those given in the previous papers allowing more general regularizations. We report on numerical simulations.