Artigo
String-averaging expectation-maximization for maximum likelihood estimation in emission tomography
Registro en:
Inverse Problems. Bristol: Iop Publishing Ltd, v. 30, n. 5, 20 p., 2014.
0266-5611
10.1088/0266-5611/30/5/055003
WOS:000336265400003
Autor
Helou, Elias Salomao [UNESP]
Censor, Yair
Chen, Tai-Been
Chern, I-Liang
De Pierro, Alvaro Rodolfo [UNESP]
Jiang, Ming
Lu, Henry Horng-Shing
Resumen
We study the maximum likelihood model in emission tomography and propose a new family of algorithms for its solution, called string-averaging expectation maximization (SAEM). In the string-averaging algorithmic regime, the index set of all underlying equations is split into subsets, called 'strings', and the algorithm separately proceeds along each string, possibly in parallel. Then, the end-points of all strings are averaged to form the next iterate. SAEM algorithms with several strings present better practical merits than the classical row-action maximum-likelihood algorithm. We present numerical experiments showing the effectiveness of the algorithmic scheme, using data of image reconstruction problems. Performance is evaluated from the computational cost and reconstruction quality viewpoints. A complete convergence theory is also provided. Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) United States-Israel Binational Science Foundation (BSF) US Department of Army award National Science Council of the Republic of China, Taiwan National Center for Theoretical Sciences (Taipei Office) National Science Council of the Republic of China Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) National Basic Research and Development Program of China (973 Program) National Science Foundation of China National Science Council National Center for Theoretical Sciences Center of Mathematical Modeling and Scientific Computing at National Chiao Tung University in Taiwan State Univ Sao Paulo, Dept Appl Math & Stat, Sao Carlos, SP, Brazil Univ Haifa, Dept Math, IL-3190501 Haifa, Israel I Shou Univ, Dept Med Imaging & Radiol Sci, Kaohsiung 82445, Taiwan Natl Chiao Tung Univ, Ctr Math Modeling & Sci Comp, Dept Appl Math, Hsinchu 30010, Taiwan Natl Taiwan Univ, Dept Math, Taipei 10617, Taiwan Peking Univ, Beijing Int Ctr Math Res, Sch Math Sci, LMAM, Beijing 100871, Peoples R China Natl Chiao Tung Univ, Inst Stat, Hsinchu 30010, Taiwan State Univ Sao Paulo, Dept Appl Math & Stat, Sao Carlos, SP, Brazil FAPESP: 13/16508-3 United States-Israel Binational Science Foundation (BSF)200912 US Department of Army awardW81XWH-10-1-0170 National Science Council of the Republic of China, TaiwanNSC 97-2118-M-214-001-MY2 National Science Council of the Republic of ChinaNSC 99-2115-M-002-003-MY3 CNPq: 301064/2009-1 National Basic Research and Development Program of China (973 Program)2011CB809105 National Science Foundation of China61121002 National Science Foundation of China10990013 National Science Foundation of China60325101