Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping

Ruru Xu; Ilkay Oksuz

doi:10.1007/978-3-031-23443-9_31

Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping

Ruru Xu^*, Ilkay Oksuz

^*Corresponding author for this work

Department of Computer Engineering

Istanbul Technical University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

Magnetic resonance imaging (MRI) is accelerated through subsampling of the associated Fourier domain in current clinical practice. The decisions on subsampling strategies and acceleration factors are provided heuristically before the acquisition. In this paper, we propose a reinforcement learning strategy for automatically deciding a subsampling strategy and acceleration factor for cardiac image acquisition. We build an environment that has a set of actions, including which k-space line to select next and when to stop the acquisition. We propose to use a reward term that penalizes extra line acquisitions and favours improved image quality. Experiments on cardiac MRI with different weightings of the reward function have shown that our method can achieve better image quality results without increasing the acquisition time and can automatically stop the k-space sampling process.

Original language	English
Title of host publication	Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers
Editors	Oscar Camara, Esther Puyol-Antón, Avan Suinesiaputra, Alistair Young, Chen Qin, Maxime Sermesant, Shuo Wang
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	340-348
Number of pages	9
ISBN (Print)	9783031234422
DOIs	https://doi.org/10.1007/978-3-031-23443-9_31
Publication status	Published - 2022
Event	13th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2022, held in conjunction with the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022 - Singapore, Singapore Duration: 18 Sept 2022 → 18 Sept 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13593 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	13th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2022, held in conjunction with the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022
Country/Territory	Singapore
City	Singapore
Period	18/09/22 → 18/09/22

Bibliographical note

Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Funding

Acknowledgments. This paper has been produced benefiting from the 2232 International Fellowship for Outstanding Researchers Program of TUBITAK (Project No: 118C353). However, the entire responsibility of the publication/paper belongs to the owner of the paper. The financial support received from TUBITAK does not mean that the content of the publication is approved in a scientific sense by TUBITAK.

Funders	Funder number
Türkiye Bilimsel ve Teknolojik Araştırma Kurumu	118C353

Keywords

Cardiac MRI
Reconstruction
Reinforcement learning

Access to Document

10.1007/978-3-031-23443-9_31

Cite this

Xu, R., & Oksuz, I. (2022). Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping. In O. Camara, E. Puyol-Antón, A. Suinesiaputra, A. Young, C. Qin, M. Sermesant, & S. Wang (Eds.), Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers (pp. 340-348). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13593 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-23443-9_31

Xu, Ruru ; Oksuz, Ilkay. / Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping. Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers. editor / Oscar Camara ; Esther Puyol-Antón ; Avan Suinesiaputra ; Alistair Young ; Chen Qin ; Maxime Sermesant ; Shuo Wang. Springer Science and Business Media Deutschland GmbH, 2022. pp. 340-348 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{960da62cad184af29644efb82e65b877,

title = "Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping",

abstract = "Magnetic resonance imaging (MRI) is accelerated through subsampling of the associated Fourier domain in current clinical practice. The decisions on subsampling strategies and acceleration factors are provided heuristically before the acquisition. In this paper, we propose a reinforcement learning strategy for automatically deciding a subsampling strategy and acceleration factor for cardiac image acquisition. We build an environment that has a set of actions, including which k-space line to select next and when to stop the acquisition. We propose to use a reward term that penalizes extra line acquisitions and favours improved image quality. Experiments on cardiac MRI with different weightings of the reward function have shown that our method can achieve better image quality results without increasing the acquisition time and can automatically stop the k-space sampling process.",

keywords = "Cardiac MRI, Reconstruction, Reinforcement learning",

author = "Ruru Xu and Ilkay Oksuz",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 13th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2022, held in conjunction with the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022 ; Conference date: 18-09-2022 Through 18-09-2022",

year = "2022",

doi = "10.1007/978-3-031-23443-9_31",

language = "English",

isbn = "9783031234422",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "340--348",

editor = "Oscar Camara and Esther Puyol-Ant{\'o}n and Avan Suinesiaputra and Alistair Young and Chen Qin and Maxime Sermesant and Shuo Wang",

booktitle = "Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers",

address = "Germany",

}

Xu, R & Oksuz, I 2022, Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping. in O Camara, E Puyol-Antón, A Suinesiaputra, A Young, C Qin, M Sermesant & S Wang (eds), Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13593 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 340-348, 13th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2022, held in conjunction with the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022, Singapore, Singapore, 18/09/22. https://doi.org/10.1007/978-3-031-23443-9_31

Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping. / Xu, Ruru; Oksuz, Ilkay.
Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers. ed. / Oscar Camara; Esther Puyol-Antón; Avan Suinesiaputra; Alistair Young; Chen Qin; Maxime Sermesant; Shuo Wang. Springer Science and Business Media Deutschland GmbH, 2022. p. 340-348 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13593 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping

AU - Xu, Ruru

AU - Oksuz, Ilkay

PY - 2022

Y1 - 2022

N2 - Magnetic resonance imaging (MRI) is accelerated through subsampling of the associated Fourier domain in current clinical practice. The decisions on subsampling strategies and acceleration factors are provided heuristically before the acquisition. In this paper, we propose a reinforcement learning strategy for automatically deciding a subsampling strategy and acceleration factor for cardiac image acquisition. We build an environment that has a set of actions, including which k-space line to select next and when to stop the acquisition. We propose to use a reward term that penalizes extra line acquisitions and favours improved image quality. Experiments on cardiac MRI with different weightings of the reward function have shown that our method can achieve better image quality results without increasing the acquisition time and can automatically stop the k-space sampling process.

AB - Magnetic resonance imaging (MRI) is accelerated through subsampling of the associated Fourier domain in current clinical practice. The decisions on subsampling strategies and acceleration factors are provided heuristically before the acquisition. In this paper, we propose a reinforcement learning strategy for automatically deciding a subsampling strategy and acceleration factor for cardiac image acquisition. We build an environment that has a set of actions, including which k-space line to select next and when to stop the acquisition. We propose to use a reward term that penalizes extra line acquisitions and favours improved image quality. Experiments on cardiac MRI with different weightings of the reward function have shown that our method can achieve better image quality results without increasing the acquisition time and can automatically stop the k-space sampling process.

KW - Cardiac MRI

KW - Reconstruction

KW - Reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=85148047982&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-23443-9_31

DO - 10.1007/978-3-031-23443-9_31

M3 - Conference contribution

AN - SCOPUS:85148047982

SN - 9783031234422

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 340

EP - 348

BT - Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers

A2 - Camara, Oscar

A2 - Puyol-Antón, Esther

A2 - Suinesiaputra, Avan

A2 - Young, Alistair

A2 - Qin, Chen

A2 - Sermesant, Maxime

A2 - Wang, Shuo

PB - Springer Science and Business Media Deutschland GmbH

T2 - 13th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2022, held in conjunction with the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022

Y2 - 18 September 2022 through 18 September 2022

ER -

Xu R, Oksuz I. Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping. In Camara O, Puyol-Antón E, Suinesiaputra A, Young A, Qin C, Sermesant M, Wang S, editors, Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers - 13th International Workshop, STACOM 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2022. p. 340-348. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-23443-9_31

Efficient MRI Reconstruction with Reinforcement Learning for Automatic Acquisition Stopping

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this