Cargando…
Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation
Iterative reconstruction has demonstrated superior performance in medical imaging under compressed, sparse, and limited-view sensing scenarios. However, iterative reconstruction algorithms are slow to converge and rely heavily on hand-crafted parameters to achieve good performance. Many iterations a...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Elsevier
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9867977/ https://www.ncbi.nlm.nih.gov/pubmed/36700132 http://dx.doi.org/10.1016/j.pacs.2023.100452 |
_version_ | 1784876450198847488 |
---|---|
author | Hsu, Ko-Tsung Guan, Steven Chitnis, Parag V. |
author_facet | Hsu, Ko-Tsung Guan, Steven Chitnis, Parag V. |
author_sort | Hsu, Ko-Tsung |
collection | PubMed |
description | Iterative reconstruction has demonstrated superior performance in medical imaging under compressed, sparse, and limited-view sensing scenarios. However, iterative reconstruction algorithms are slow to converge and rely heavily on hand-crafted parameters to achieve good performance. Many iterations are usually required to reconstruct a high-quality image, which is computationally expensive due to repeated evaluations of the physical model. While learned iterative reconstruction approaches such as model-based learning (MBLr) can reduce the number of iterations through convolutional neural networks, it still requires repeated evaluations of the physical models at each iteration. Therefore, the goal of this study is to develop a Fast Iterative Reconstruction (FIRe) algorithm that incorporates a learned physical model into the learned iterative reconstruction scheme to further reduce the reconstruction time while maintaining robust reconstruction performance. We also propose an efficient training scheme for FIRe, which releases the enormous memory footprint required by learned iterative reconstruction methods through the concept of recursive training. The results of our proposed method demonstrate comparable reconstruction performance to learned iterative reconstruction methods with a 9x reduction in computation time and a 620x reduction in computation time compared to variational reconstruction. |
format | Online Article Text |
id | pubmed-9867977 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | Elsevier |
record_format | MEDLINE/PubMed |
spelling | pubmed-98679772023-01-24 Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation Hsu, Ko-Tsung Guan, Steven Chitnis, Parag V. Photoacoustics Research Article Iterative reconstruction has demonstrated superior performance in medical imaging under compressed, sparse, and limited-view sensing scenarios. However, iterative reconstruction algorithms are slow to converge and rely heavily on hand-crafted parameters to achieve good performance. Many iterations are usually required to reconstruct a high-quality image, which is computationally expensive due to repeated evaluations of the physical model. While learned iterative reconstruction approaches such as model-based learning (MBLr) can reduce the number of iterations through convolutional neural networks, it still requires repeated evaluations of the physical models at each iteration. Therefore, the goal of this study is to develop a Fast Iterative Reconstruction (FIRe) algorithm that incorporates a learned physical model into the learned iterative reconstruction scheme to further reduce the reconstruction time while maintaining robust reconstruction performance. We also propose an efficient training scheme for FIRe, which releases the enormous memory footprint required by learned iterative reconstruction methods through the concept of recursive training. The results of our proposed method demonstrate comparable reconstruction performance to learned iterative reconstruction methods with a 9x reduction in computation time and a 620x reduction in computation time compared to variational reconstruction. Elsevier 2023-01-13 /pmc/articles/PMC9867977/ /pubmed/36700132 http://dx.doi.org/10.1016/j.pacs.2023.100452 Text en © 2023 The Authors https://creativecommons.org/licenses/by-nc-nd/4.0/This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). |
spellingShingle | Research Article Hsu, Ko-Tsung Guan, Steven Chitnis, Parag V. Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation |
title | Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation |
title_full | Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation |
title_fullStr | Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation |
title_full_unstemmed | Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation |
title_short | Fast iterative reconstruction for photoacoustic tomography using learned physical model: Theoretical validation |
title_sort | fast iterative reconstruction for photoacoustic tomography using learned physical model: theoretical validation |
topic | Research Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9867977/ https://www.ncbi.nlm.nih.gov/pubmed/36700132 http://dx.doi.org/10.1016/j.pacs.2023.100452 |
work_keys_str_mv | AT hsukotsung fastiterativereconstructionforphotoacoustictomographyusinglearnedphysicalmodeltheoreticalvalidation AT guansteven fastiterativereconstructionforphotoacoustictomographyusinglearnedphysicalmodeltheoreticalvalidation AT chitnisparagv fastiterativereconstructionforphotoacoustictomographyusinglearnedphysicalmodeltheoreticalvalidation |