Monotone FISTA With Variable Acceleration for Compressed Sensing Magnetic Resonance Imaging
| Autor: | Marcelo V. W. Zibetti, Gabor T. Herman, Elias S. Helou, Ravinder R. Regatte |
|---|---|
| Přispěvatelé: | New York Univ, Universidade Estadual Paulista (Unesp), CUNY |
| Rok vydání: | 2019 |
| Předmět: |
iterative algorithms
Computer science Computation ALGORITMOS PARA IMAGENS Context (language use) 02 engineering and technology Iterative reconstruction Article 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Convergence (routing) 0202 electrical engineering electronic engineering information engineering magnetic resonance imaging Proximal-gradient methods FISTA compressed sensing Line search Computer Science Applications Computational Mathematics Monotone polygon Compressed sensing Signal Processing 020201 artificial intelligence & image processing Proximal Gradient Methods Algorithm |
| Zdroj: | Web of Science Repositório Institucional da UNESP Universidade Estadual Paulista (UNESP) instacron:UNESP Repositório Institucional da USP (Biblioteca Digital da Produção Intelectual) Universidade de São Paulo (USP) instacron:USP |
| ISSN: | 2334-0118 2573-0436 |
| DOI: | 10.1109/tci.2018.2882681 |
| Popis: | Made available in DSpace on 2019-10-04T11:56:54Z (GMT). No. of bitstreams: 0 Previous issue date: 2019-03-01 NIH Center of Advanced Imaging Innovation and Research (CAI2R) NIBIB Biomedical Technology Resource Center Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) An improvement of the monotone fast iterative shrinkage-thresholding algorithm (MFISTA) for faster convergence is proposed in this paper. Our motivation is to reduce the reconstruction time of compressed sensing problems in magnetic resonance imaging. The proposed modification introduces an extra term, which is a multiple of the proximal-gradient step, into the so-called momentum formula used for the computation of the next iterate in MFISTA. In addition, the modified algorithm selects the next iterate as a possibly improved point obtained by any other procedure, such as an arbitrary shift, a line search, or other methods. As an example, an arbitrary-length shift in the direction from the previous iterate to the output of the proximal-gradient step is considered. The resulting algorithm accelerates MFISTA in a manner that varies with the iterative steps. Convergence analysis shows that the proposed modification provides improved theoretical convergence bounds, and that it has more flexibility in its parameters than the original MFISTA. Since such problems need to he studied in the context of functions of several complex variables, a careful extension of FISTA-like methods to complex variables is provided. New York Univ, Sch Med, New York, NY 10016 USA State Univ Sao Paulo, BR-01049010 Sao Paulo, Brazil CUNY, New York, NY 10017 USA State Univ Sao Paulo, BR-01049010 Sao Paulo, Brazil NIH: R01-AR060238 NIH: R01-AR067156 NIH: R01-AR068966 NIBIB Biomedical Technology Resource Center: NIH P41-EB017183 FAPESP: 2013/07375-0 FAPESP: 2016/24286-9 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|