期刊
IEEE TRANSACTIONS ON MEDICAL IMAGING
卷 37, 期 4, 页码 1000-1010出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMI.2017.2786865
关键词
Emission tomography; penalized reconstruction; L-BFGS-B; preconditioning
类别
资金
- GE Healthcare
- NIHR
- Leverhulme Trust
- EPSRC [EP/M00483X/1, EP/N014588/1]
- Cantab Capital Institute for the Mathematics of Information
- CHiPS (Horizon 2020 RISE Project Grant)
- Engineering and Physical Sciences Research Council [EP/N014588/1, EP/E034950/1, EP/M00483X/1, EP/K005278/1] Funding Source: researchfish
- EPSRC [EP/K005278/1, EP/M022587/1, EP/E034950/1] Funding Source: UKRI
This paper reports on the feasibility of using a quasi-Newton optimization algorithm, limited-memory Broyden-Fletcher-Goldfarb-Shanno with boundary constraints (L-BFGS-B), for penalized image reconstruction problems in emission tomography (ET). For further acceleration, an additional preconditioning technique based on a diagonal approximation of the Hessian was introduced. The convergence rate of L-BFGS-B and the proposed pre-conditioned algorithm (L-BFGS-B-PC) was evaluated with simulated data with various factors, such as the noise level, penalty type, penalty strength and background level. Data of three F-18-FDG patient acquisitions were also reconstructed. Results showed that the proposed L-BFGS-B-PC outperforms L-BFGS-B in convergence rate for all simulated conditions and the patient data. Based on these results, L-BFGS-B-PC shows promise for clinical application.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据