AI-Based Reconstruction for Fast MRI-A Systematic Review and Meta-Analysis

Yutong Chen; Carola Bibiane Schonlieb; Pietro Lio; Tim Leiner; Pier Luigi Dragotti; Ge Wang; Daniel Rueckert; David Firmin; Guang Yang

doi:10.1109/JPROC.2022.3141367

AI-Based Reconstruction for Fast MRI-A Systematic Review and Meta-Analysis

Yutong Chen, Carola Bibiane Schonlieb, Pietro Lio, Tim Leiner, Pier Luigi Dragotti, Ge Wang, Daniel Rueckert, David Firmin, Guang Yang

Research output: Contribution to journal › Review article › peer-review

83 Scopus citations

Abstract

Compressed sensing (CS) has been playing a key role in accelerating the magnetic resonance imaging (MRI) acquisition process. With the resurgence of artificial intelligence, deep neural networks and CS algorithms are being integrated to redefine the state of the art of fast MRI. The past several years have witnessed substantial growth in the complexity, diversity, and performance of deep-learning-based CS techniques that are dedicated to fast MRI. In this meta-analysis, we systematically review the deep-learning-based CS techniques for fast MRI, describe key model designs, highlight breakthroughs, and discuss promising directions. We have also introduced a comprehensive analysis framework and a classification system to assess the pivotal role of deep learning in CS-based acceleration for MRI.

Original language	English
Pages (from-to)	224-245
Number of pages	22
Journal	Proceedings of the IEEE
Volume	110
Issue number	2
DOIs	https://doi.org/10.1109/JPROC.2022.3141367
State	Published - 1 Feb 2022
Externally published	Yes

Keywords

Compressed sensing (CS)
Deep learning
Magnetic resonance imaging (MRI)
Neural network

Access to Document

10.1109/JPROC.2022.3141367

Cite this

@article{6d045233901648e09d36499acc71aa8a,

title = "AI-Based Reconstruction for Fast MRI-A Systematic Review and Meta-Analysis",

abstract = "Compressed sensing (CS) has been playing a key role in accelerating the magnetic resonance imaging (MRI) acquisition process. With the resurgence of artificial intelligence, deep neural networks and CS algorithms are being integrated to redefine the state of the art of fast MRI. The past several years have witnessed substantial growth in the complexity, diversity, and performance of deep-learning-based CS techniques that are dedicated to fast MRI. In this meta-analysis, we systematically review the deep-learning-based CS techniques for fast MRI, describe key model designs, highlight breakthroughs, and discuss promising directions. We have also introduced a comprehensive analysis framework and a classification system to assess the pivotal role of deep learning in CS-based acceleration for MRI.",

keywords = "Compressed sensing (CS), Deep learning, Magnetic resonance imaging (MRI), Neural network",

author = "Yutong Chen and Schonlieb, {Carola Bibiane} and Pietro Lio and Tim Leiner and Dragotti, {Pier Luigi} and Ge Wang and Daniel Rueckert and David Firmin and Guang Yang",

note = "Publisher Copyright: {\textcopyright} 1963-2012 IEEE.",

year = "2022",

month = feb,

day = "1",

doi = "10.1109/JPROC.2022.3141367",

language = "English",

volume = "110",

pages = "224--245",

journal = "Proceedings of the IEEE",

issn = "0018-9219",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - AI-Based Reconstruction for Fast MRI-A Systematic Review and Meta-Analysis

AU - Chen, Yutong

AU - Schonlieb, Carola Bibiane

AU - Lio, Pietro

AU - Leiner, Tim

AU - Dragotti, Pier Luigi

AU - Wang, Ge

AU - Rueckert, Daniel

AU - Firmin, David

AU - Yang, Guang

PY - 2022/2/1

Y1 - 2022/2/1

N2 - Compressed sensing (CS) has been playing a key role in accelerating the magnetic resonance imaging (MRI) acquisition process. With the resurgence of artificial intelligence, deep neural networks and CS algorithms are being integrated to redefine the state of the art of fast MRI. The past several years have witnessed substantial growth in the complexity, diversity, and performance of deep-learning-based CS techniques that are dedicated to fast MRI. In this meta-analysis, we systematically review the deep-learning-based CS techniques for fast MRI, describe key model designs, highlight breakthroughs, and discuss promising directions. We have also introduced a comprehensive analysis framework and a classification system to assess the pivotal role of deep learning in CS-based acceleration for MRI.

AB - Compressed sensing (CS) has been playing a key role in accelerating the magnetic resonance imaging (MRI) acquisition process. With the resurgence of artificial intelligence, deep neural networks and CS algorithms are being integrated to redefine the state of the art of fast MRI. The past several years have witnessed substantial growth in the complexity, diversity, and performance of deep-learning-based CS techniques that are dedicated to fast MRI. In this meta-analysis, we systematically review the deep-learning-based CS techniques for fast MRI, describe key model designs, highlight breakthroughs, and discuss promising directions. We have also introduced a comprehensive analysis framework and a classification system to assess the pivotal role of deep learning in CS-based acceleration for MRI.

KW - Compressed sensing (CS)

KW - Deep learning

KW - Magnetic resonance imaging (MRI)

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=85124598976&partnerID=8YFLogxK

U2 - 10.1109/JPROC.2022.3141367

DO - 10.1109/JPROC.2022.3141367

M3 - Review article

AN - SCOPUS:85124598976

SN - 0018-9219

VL - 110

SP - 224

EP - 245

JO - Proceedings of the IEEE

JF - Proceedings of the IEEE

IS - 2

ER -

AI-Based Reconstruction for Fast MRI-A Systematic Review and Meta-Analysis

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this