Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance

Dewmini Hasara Wickremasinghe; Natallia Khenkina; Pier Giorgio Masci; Andrew P. King; Esther Puyol-Antón

doi:10.1007/978-3-030-80432-9_8

Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance

Dewmini Hasara Wickremasinghe^*, Natallia Khenkina, Pier Giorgio Masci, Andrew P. King, Esther Puyol-Antón

^*Corresponding author for this work

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

Abstract

Cardiovascular magnetic resonance (CMR) is an established, non-invasive technique to comprehensively assess cardiovascular structure and function in a variety of acquired and inherited cardiac conditions. In addition to the heart, a typical CMR examination will also image adjacent thoracic and abdominal structures. Consequently, findings incidental to the cardiac examination may be encountered, some of which may be clinically relevant. We compare two deep learning architectures to automatically detect extra cardiac findings (ECFs) in the HASTE sequence of a CMR acquisition. The first one consists of a binary classification network that detects the presence of ECFs and the second one is a multi-label classification network that detects and classifies the type of ECF. We validated the two models on a cohort of 236 subjects, corresponding to 5610 slices, where 746 ECFs were found. Results show that the proposed methods have promising balanced accuracy and sensitivity and high specificity.

Original language	English
Title of host publication	Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings
Editors	Bartłomiej W. Papież, Mohammad Yaqub, Jianbo Jiao, Ana I. Namburete, J. Alison Noble
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	98-107
Number of pages	10
ISBN (Print)	9783030804312
DOIs	https://doi.org/10.1007/978-3-030-80432-9_8
Publication status	Published - 2021
Event	25th Annual Conference on Medical Image Understanding and Analysis, MIUA 2021 - Virtual, Online Duration: 12 Jul 2021 → 14 Jul 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12722 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th Annual Conference on Medical Image Understanding and Analysis, MIUA 2021
City	Virtual, Online
Period	12/07/2021 → 14/07/2021

Keywords

Cardiovascular magnetic resonance
Deep learning classification
Extra cardiac findings

Access to Document

10.1007/978-3-030-80432-9_8

Cite this

Wickremasinghe, D. H., Khenkina, N., Masci, P. G., King, A. P., & Puyol-Antón, E. (2021). Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance. In B. W. Papież, M. Yaqub, J. Jiao, A. I. Namburete, & J. A. Noble (Eds.), Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings (pp. 98-107). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12722 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-80432-9_8

Wickremasinghe, Dewmini Hasara ; Khenkina, Natallia ; Masci, Pier Giorgio et al. / Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance. Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings. editor / Bartłomiej W. Papież ; Mohammad Yaqub ; Jianbo Jiao ; Ana I. Namburete ; J. Alison Noble. Springer Science and Business Media Deutschland GmbH, 2021. pp. 98-107 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inbook{5b66608b42334845ad842ca142e5d30e,

title = "Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance",

abstract = "Cardiovascular magnetic resonance (CMR) is an established, non-invasive technique to comprehensively assess cardiovascular structure and function in a variety of acquired and inherited cardiac conditions. In addition to the heart, a typical CMR examination will also image adjacent thoracic and abdominal structures. Consequently, findings incidental to the cardiac examination may be encountered, some of which may be clinically relevant. We compare two deep learning architectures to automatically detect extra cardiac findings (ECFs) in the HASTE sequence of a CMR acquisition. The first one consists of a binary classification network that detects the presence of ECFs and the second one is a multi-label classification network that detects and classifies the type of ECF. We validated the two models on a cohort of 236 subjects, corresponding to 5610 slices, where 746 ECFs were found. Results show that the proposed methods have promising balanced accuracy and sensitivity and high specificity.",

keywords = "Cardiovascular magnetic resonance, Deep learning classification, Extra cardiac findings",

author = "Wickremasinghe, {Dewmini Hasara} and Natallia Khenkina and Masci, {Pier Giorgio} and King, {Andrew P.} and Esther Puyol-Ant{\'o}n",

note = "Funding Information: This work was supported by the EPSRC (EP/R005516/1 and EP/P001009/1) and the Wellcome EPSRC Centre for Medical Engineering at the School of Biomedical Engineering and Imaging Sciences, King?s College London (WT 203148/Z/16/Z). Funding Information: Acknowledgements. This work was supported by the EPSRC (EP/R005516/1 and EP/P001009/1) and the Wellcome EPSRC Centre for Medical Engineering at the School of Biomedical Engineering and Imaging Sciences, King{\textquoteright}s College London (WT 203148/Z/16/Z). Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.; 25th Annual Conference on Medical Image Understanding and Analysis, MIUA 2021 ; Conference date: 12-07-2021 Through 14-07-2021",

year = "2021",

doi = "10.1007/978-3-030-80432-9_8",

language = "English",

isbn = "9783030804312",

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 = "98--107",

editor = "Papie{\.z}, {Bart{\l}omiej W.} and Mohammad Yaqub and Jianbo Jiao and Namburete, {Ana I.} and Noble, {J. Alison}",

booktitle = "Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings",

address = "Germany",

}

Wickremasinghe, DH, Khenkina, N, Masci, PG , King, AP & Puyol-Antón, E 2021, Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance. in BW Papież, M Yaqub, J Jiao, AI Namburete & JA Noble (eds), Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12722 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 98-107, 25th Annual Conference on Medical Image Understanding and Analysis, MIUA 2021, Virtual, Online, 12/07/2021. https://doi.org/10.1007/978-3-030-80432-9_8

Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance. / Wickremasinghe, Dewmini Hasara; Khenkina, Natallia; Masci, Pier Giorgio et al.
Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings. ed. / Bartłomiej W. Papież; Mohammad Yaqub; Jianbo Jiao; Ana I. Namburete; J. Alison Noble. Springer Science and Business Media Deutschland GmbH, 2021. p. 98-107 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12722 LNCS).

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

TY - CHAP

T1 - Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance

AU - Wickremasinghe, Dewmini Hasara

AU - Khenkina, Natallia

AU - Masci, Pier Giorgio

AU - King, Andrew P.

AU - Puyol-Antón, Esther

N1 - Funding Information: This work was supported by the EPSRC (EP/R005516/1 and EP/P001009/1) and the Wellcome EPSRC Centre for Medical Engineering at the School of Biomedical Engineering and Imaging Sciences, King?s College London (WT 203148/Z/16/Z). Funding Information: Acknowledgements. This work was supported by the EPSRC (EP/R005516/1 and EP/P001009/1) and the Wellcome EPSRC Centre for Medical Engineering at the School of Biomedical Engineering and Imaging Sciences, King’s College London (WT 203148/Z/16/Z). Publisher Copyright: © 2021, Springer Nature Switzerland AG. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021

Y1 - 2021

N2 - Cardiovascular magnetic resonance (CMR) is an established, non-invasive technique to comprehensively assess cardiovascular structure and function in a variety of acquired and inherited cardiac conditions. In addition to the heart, a typical CMR examination will also image adjacent thoracic and abdominal structures. Consequently, findings incidental to the cardiac examination may be encountered, some of which may be clinically relevant. We compare two deep learning architectures to automatically detect extra cardiac findings (ECFs) in the HASTE sequence of a CMR acquisition. The first one consists of a binary classification network that detects the presence of ECFs and the second one is a multi-label classification network that detects and classifies the type of ECF. We validated the two models on a cohort of 236 subjects, corresponding to 5610 slices, where 746 ECFs were found. Results show that the proposed methods have promising balanced accuracy and sensitivity and high specificity.

AB - Cardiovascular magnetic resonance (CMR) is an established, non-invasive technique to comprehensively assess cardiovascular structure and function in a variety of acquired and inherited cardiac conditions. In addition to the heart, a typical CMR examination will also image adjacent thoracic and abdominal structures. Consequently, findings incidental to the cardiac examination may be encountered, some of which may be clinically relevant. We compare two deep learning architectures to automatically detect extra cardiac findings (ECFs) in the HASTE sequence of a CMR acquisition. The first one consists of a binary classification network that detects the presence of ECFs and the second one is a multi-label classification network that detects and classifies the type of ECF. We validated the two models on a cohort of 236 subjects, corresponding to 5610 slices, where 746 ECFs were found. Results show that the proposed methods have promising balanced accuracy and sensitivity and high specificity.

KW - Cardiovascular magnetic resonance

KW - Deep learning classification

KW - Extra cardiac findings

UR - http://www.scopus.com/inward/record.url?scp=85112198329&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-80432-9_8

DO - 10.1007/978-3-030-80432-9_8

M3 - Conference paper

AN - SCOPUS:85112198329

SN - 9783030804312

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

SP - 98

EP - 107

BT - Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings

A2 - Papież, Bartłomiej W.

A2 - Yaqub, Mohammad

A2 - Jiao, Jianbo

A2 - Namburete, Ana I.

A2 - Noble, J. Alison

PB - Springer Science and Business Media Deutschland GmbH

T2 - 25th Annual Conference on Medical Image Understanding and Analysis, MIUA 2021

Y2 - 12 July 2021 through 14 July 2021

ER -

Wickremasinghe DH, Khenkina N, Masci PG , King AP, Puyol-Antón E. Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance. In Papież BW, Yaqub M, Jiao J, Namburete AI, Noble JA, editors, Medical Image Understanding and Analysis - 25th Annual Conference, MIUA 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 98-107. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-80432-9_8

Automatic Detection of Extra-Cardiac Findings in Cardiovascular Magnetic Resonance

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this