An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset

This work aims at developing a generalizable Magnetic Resonance Imaging (MRI) reconstruction method in the meta-learning framework. Specifically, we develop a deep reconstruction network induced by a learnable optimization algorithm (LOA) to solve the nonconvex nonsmooth variational model of MRI ima...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Wanyu Bian, Yunmei Chen, Xiaojing Ye, Qingchao Zhang
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	MRI reconstruction meta-learning domain generalization Photography TR1-1050 Computer applications to medicine. Medical informatics R858-859.7 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/6ecd95b04f2147f4b78df9f402a73dea
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:6ecd95b04f2147f4b78df9f402a73dea
record_format	dspace
spelling	oai:doaj.org-article:6ecd95b04f2147f4b78df9f402a73dea2021-11-25T18:03:28ZAn Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset10.3390/jimaging71102312313-433Xhttps://doaj.org/article/6ecd95b04f2147f4b78df9f402a73dea2021-10-01T00:00:00Zhttps://www.mdpi.com/2313-433X/7/11/231https://doaj.org/toc/2313-433XThis work aims at developing a generalizable Magnetic Resonance Imaging (MRI) reconstruction method in the meta-learning framework. Specifically, we develop a deep reconstruction network induced by a learnable optimization algorithm (LOA) to solve the nonconvex nonsmooth variational model of MRI image reconstruction. In this model, the nonconvex nonsmooth regularization term is parameterized as a structured deep network where the network parameters can be learned from data. We partition these network parameters into two parts: a task-invariant part for the common feature encoder component of the regularization, and a task-specific part to account for the variations in the heterogeneous training and testing data. We train the regularization parameters in a bilevel optimization framework which significantly improves the robustness of the training process and the generalization ability of the network. We conduct a series of numerical experiments using heterogeneous MRI data sets with various undersampling patterns, ratios, and acquisition settings. The experimental results show that our network yields greatly improved reconstruction quality over existing methods and can generalize well to new reconstruction problems whose undersampling patterns/trajectories are not present during training.Wanyu BianYunmei ChenXiaojing YeQingchao ZhangMDPI AGarticleMRI reconstructionmeta-learningdomain generalizationPhotographyTR1-1050Computer applications to medicine. Medical informaticsR858-859.7Electronic computers. Computer scienceQA75.5-76.95ENJournal of Imaging, Vol 7, Iss 231, p 231 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	MRI reconstruction meta-learning domain generalization Photography TR1-1050 Computer applications to medicine. Medical informatics R858-859.7 Electronic computers. Computer science QA75.5-76.95
spellingShingle	MRI reconstruction meta-learning domain generalization Photography TR1-1050 Computer applications to medicine. Medical informatics R858-859.7 Electronic computers. Computer science QA75.5-76.95 Wanyu Bian Yunmei Chen Xiaojing Ye Qingchao Zhang An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset
description	This work aims at developing a generalizable Magnetic Resonance Imaging (MRI) reconstruction method in the meta-learning framework. Specifically, we develop a deep reconstruction network induced by a learnable optimization algorithm (LOA) to solve the nonconvex nonsmooth variational model of MRI image reconstruction. In this model, the nonconvex nonsmooth regularization term is parameterized as a structured deep network where the network parameters can be learned from data. We partition these network parameters into two parts: a task-invariant part for the common feature encoder component of the regularization, and a task-specific part to account for the variations in the heterogeneous training and testing data. We train the regularization parameters in a bilevel optimization framework which significantly improves the robustness of the training process and the generalization ability of the network. We conduct a series of numerical experiments using heterogeneous MRI data sets with various undersampling patterns, ratios, and acquisition settings. The experimental results show that our network yields greatly improved reconstruction quality over existing methods and can generalize well to new reconstruction problems whose undersampling patterns/trajectories are not present during training.
format	article
author	Wanyu Bian Yunmei Chen Xiaojing Ye Qingchao Zhang
author_facet	Wanyu Bian Yunmei Chen Xiaojing Ye Qingchao Zhang
author_sort	Wanyu Bian
title	An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset
title_short	An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset
title_full	An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset
title_fullStr	An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset
title_full_unstemmed	An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset
title_sort	optimization-based meta-learning model for mri reconstruction with diverse dataset
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/6ecd95b04f2147f4b78df9f402a73dea
work_keys_str_mv	AT wanyubian anoptimizationbasedmetalearningmodelformrireconstructionwithdiversedataset AT yunmeichen anoptimizationbasedmetalearningmodelformrireconstructionwithdiversedataset AT xiaojingye anoptimizationbasedmetalearningmodelformrireconstructionwithdiversedataset AT qingchaozhang anoptimizationbasedmetalearningmodelformrireconstructionwithdiversedataset AT wanyubian optimizationbasedmetalearningmodelformrireconstructionwithdiversedataset AT yunmeichen optimizationbasedmetalearningmodelformrireconstructionwithdiversedataset AT xiaojingye optimizationbasedmetalearningmodelformrireconstructionwithdiversedataset AT qingchaozhang optimizationbasedmetalearningmodelformrireconstructionwithdiversedataset
_version_	1718411684631019520

An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset

Ejemplares similares