Development of iterative algorithms for image reconstruction

The iterative approach is important for computed tomography (CT) and attracting more and more attention with the rapid evolution of computer technology. In this paper, classic results and recent advances on iterative algorithms for image reconstruction are reviewed, with an emphasis on the ART-like and EM-like algorithms in both of their simultaneous and ordered-subset formats. The following issues are discussed: what the computational structures are, under what conditions the algorithms converge, what the final limits are, what the relaxation strategies we have, how the final limits depend on the initial guesses, and so on. In addition, heuristic arguments are given for the SART and EM algorithms for understanding the algorithms. Finally, future research directions are discussed, along with guidelines for practical applications of iterative algorithms.Instruments & InstrumentationOpticsPhysics, AppliedSCI(E)0ARTICLE1-277-861