Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels

Authors

Yu-Jung Tsai, University College London
Alexandre B. Bousse, University College London
Matthias J. Ehrhardt, University College London
Brian F. Hutton, University of WollongongFollow
Simon R. Arridge, University College London
Kris Thielemans, University College London

RIS ID

110534

Publication Details

Tsai, Y., Bousse, A., Ehrhardt, M. J., Hutton, B. F., Arridge, S. & Thielemans, K. (2015). Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels. 2015 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC) (pp. 1-3). United States: IEEE.

Abstract

A new algorithm (LBFGS-B-PC) which combines ideas of two existing convergent reconstruction algorithms, relaxed separable paraboloidal surrogate (SPS) and limited-memory Broyden-Fletcher-Goldfarb-Shanno with boundary constraints (LBFGS-B), is proposed. Its performance is evaluated in terms of log-posterior value and regional recovery ratio. The results demonstrate the superior convergence speed of the proposed algorithm to relaxed SPS and LBFGS-B, regardless of the noise level, activity distribution, object geometry, and penalties.