Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence

Ross Upton; Angela Mumith; Arian Beqiri; Andrew Parker; William Hawkes; Shan Gao; Mihaela Porumb; Rizwan Sarwar; Patricia Marques; Deborah Markham; Jake Kenworthy; Jamie M O'Driscoll; Neelam Hassanali; Kate Groves; Cameron Dockerill; William Woodward; Maryam Alsharqi; Annabelle McCourt; Edmund H Wilkes; Stephen B Heitner; Mrinal Yadava; David Stojanovski; Pablo Lamata; Gary Woodward; Paul Leeson

doi:10.1016/j.jcmg.2021.10.013

Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence

Ross Upton, Angela Mumith, Arian Beqiri, Andrew Parker, William Hawkes, Shan Gao, Mihaela Porumb, Rizwan Sarwar, Patricia Marques, Deborah Markham, Jake Kenworthy, Jamie M O'Driscoll, Neelam Hassanali, Kate Groves, Cameron Dockerill, William Woodward, Maryam Alsharqi, Annabelle McCourt, Edmund H Wilkes, Stephen B HeitnerMrinal Yadava, David Stojanovski, Pablo Lamata, Gary Woodward, Paul Leeson

Biomedical Engineering Department

Research output: Contribution to journal › Article › peer-review

52 Citations (Scopus)

Abstract

OBJECTIVES: The purpose of this study was to establish whether an artificially intelligent (AI) system can be developed to automate stress echocardiography analysis and support clinician interpretation.

BACKGROUND: Coronary artery disease is the leading global cause of mortality and morbidity and stress echocardiography remains one of the most commonly used diagnostic imaging tests.

METHODS: An automated image processing pipeline was developed to extract novel geometric and kinematic features from stress echocardiograms collected as part of a large, United Kingdom-based prospective, multicenter, multivendor study. An ensemble machine learning classifier was trained, using the extracted features, to identify patients with severe coronary artery disease on invasive coronary angiography. The model was tested in an independent U.S.

STUDY: How availability of an AI classification might impact clinical interpretation of stress echocardiograms was evaluated in a randomized crossover reader study.

RESULTS: Acceptable classification accuracy for identification of patients with severe coronary artery disease in the training data set was achieved on cross-fold validation based on 31 unique geometric and kinematic features, with a specificity of 92.7% and a sensitivity of 84.4%. This accuracy was maintained in the independent validation data set. The use of the AI classification tool by clinicians increased inter-reader agreement and confidence as well as sensitivity for detection of disease by 10% to achieve an area under the receiver-operating characteristic curve of 0.93.

CONCLUSIONS: Automated analysis of stress echocardiograms is possible using AI and provision of automated classifications to clinicians when reading stress echocardiograms could improve accuracy, inter-reader agreement, and reader confidence.

Original language	English
Pages (from-to)	715-727
Number of pages	13
Journal	JACC. Cardiovascular imaging
Volume	15
Issue number	5
Early online date	2 May 2022
DOIs	https://doi.org/10.1016/j.jcmg.2021.10.013
Publication status	Published - May 2022

Keywords

Artificial Intelligence
Coronary Artery Disease/diagnostic imaging
Echocardiography/methods
Humans
Predictive Value of Tests
Prospective Studies

Access to Document

10.1016/j.jcmg.2021.10.013

Cite this

Upton, R., Mumith, A., Beqiri, A., Parker, A., Hawkes, W., Gao, S., Porumb, M., Sarwar, R., Marques, P., Markham, D., Kenworthy, J., O'Driscoll, J. M., Hassanali, N., Groves, K., Dockerill, C., Woodward, W., Alsharqi, M., McCourt, A., Wilkes, E. H., ... Leeson, P. (2022). Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence. JACC. Cardiovascular imaging, 15(5), 715-727. https://doi.org/10.1016/j.jcmg.2021.10.013

@article{e27d79d7b2d44bd890e3a4dca3339564,

title = "Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence",

abstract = "OBJECTIVES: The purpose of this study was to establish whether an artificially intelligent (AI) system can be developed to automate stress echocardiography analysis and support clinician interpretation.BACKGROUND: Coronary artery disease is the leading global cause of mortality and morbidity and stress echocardiography remains one of the most commonly used diagnostic imaging tests.METHODS: An automated image processing pipeline was developed to extract novel geometric and kinematic features from stress echocardiograms collected as part of a large, United Kingdom-based prospective, multicenter, multivendor study. An ensemble machine learning classifier was trained, using the extracted features, to identify patients with severe coronary artery disease on invasive coronary angiography. The model was tested in an independent U.S.STUDY: How availability of an AI classification might impact clinical interpretation of stress echocardiograms was evaluated in a randomized crossover reader study.RESULTS: Acceptable classification accuracy for identification of patients with severe coronary artery disease in the training data set was achieved on cross-fold validation based on 31 unique geometric and kinematic features, with a specificity of 92.7% and a sensitivity of 84.4%. This accuracy was maintained in the independent validation data set. The use of the AI classification tool by clinicians increased inter-reader agreement and confidence as well as sensitivity for detection of disease by 10% to achieve an area under the receiver-operating characteristic curve of 0.93.CONCLUSIONS: Automated analysis of stress echocardiograms is possible using AI and provision of automated classifications to clinicians when reading stress echocardiograms could improve accuracy, inter-reader agreement, and reader confidence.",

keywords = "Artificial Intelligence, Coronary Artery Disease/diagnostic imaging, Echocardiography/methods, Humans, Predictive Value of Tests, Prospective Studies",

author = "Ross Upton and Angela Mumith and Arian Beqiri and Andrew Parker and William Hawkes and Shan Gao and Mihaela Porumb and Rizwan Sarwar and Patricia Marques and Deborah Markham and Jake Kenworthy and O'Driscoll, {Jamie M} and Neelam Hassanali and Kate Groves and Cameron Dockerill and William Woodward and Maryam Alsharqi and Annabelle McCourt and Wilkes, {Edmund H} and Heitner, {Stephen B} and Mrinal Yadava and David Stojanovski and Pablo Lamata and Gary Woodward and Paul Leeson",

year = "2022",

month = may,

doi = "10.1016/j.jcmg.2021.10.013",

language = "English",

volume = "15",

pages = "715--727",

journal = "JACC. Cardiovascular imaging",

issn = "1936-878X",

publisher = "Elsevier",

number = "5",

}

Upton, R, Mumith, A, Beqiri, A, Parker, A, Hawkes, W, Gao, S, Porumb, M, Sarwar, R, Marques, P, Markham, D, Kenworthy, J, O'Driscoll, JM, Hassanali, N, Groves, K, Dockerill, C, Woodward, W, Alsharqi, M, McCourt, A, Wilkes, EH, Heitner, SB, Yadava, M, Stojanovski, D, Lamata, P, Woodward, G & Leeson, P 2022, 'Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence', JACC. Cardiovascular imaging, vol. 15, no. 5, pp. 715-727. https://doi.org/10.1016/j.jcmg.2021.10.013

TY - JOUR

T1 - Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence

AU - Upton, Ross

AU - Mumith, Angela

AU - Beqiri, Arian

AU - Parker, Andrew

AU - Hawkes, William

AU - Gao, Shan

AU - Porumb, Mihaela

AU - Sarwar, Rizwan

AU - Marques, Patricia

AU - Markham, Deborah

AU - Kenworthy, Jake

AU - O'Driscoll, Jamie M

AU - Hassanali, Neelam

AU - Groves, Kate

AU - Dockerill, Cameron

AU - Woodward, William

AU - Alsharqi, Maryam

AU - McCourt, Annabelle

AU - Wilkes, Edmund H

AU - Heitner, Stephen B

AU - Yadava, Mrinal

AU - Stojanovski, David

AU - Lamata, Pablo

AU - Woodward, Gary

AU - Leeson, Paul

PY - 2022/5

Y1 - 2022/5

N2 - OBJECTIVES: The purpose of this study was to establish whether an artificially intelligent (AI) system can be developed to automate stress echocardiography analysis and support clinician interpretation.BACKGROUND: Coronary artery disease is the leading global cause of mortality and morbidity and stress echocardiography remains one of the most commonly used diagnostic imaging tests.METHODS: An automated image processing pipeline was developed to extract novel geometric and kinematic features from stress echocardiograms collected as part of a large, United Kingdom-based prospective, multicenter, multivendor study. An ensemble machine learning classifier was trained, using the extracted features, to identify patients with severe coronary artery disease on invasive coronary angiography. The model was tested in an independent U.S.STUDY: How availability of an AI classification might impact clinical interpretation of stress echocardiograms was evaluated in a randomized crossover reader study.RESULTS: Acceptable classification accuracy for identification of patients with severe coronary artery disease in the training data set was achieved on cross-fold validation based on 31 unique geometric and kinematic features, with a specificity of 92.7% and a sensitivity of 84.4%. This accuracy was maintained in the independent validation data set. The use of the AI classification tool by clinicians increased inter-reader agreement and confidence as well as sensitivity for detection of disease by 10% to achieve an area under the receiver-operating characteristic curve of 0.93.CONCLUSIONS: Automated analysis of stress echocardiograms is possible using AI and provision of automated classifications to clinicians when reading stress echocardiograms could improve accuracy, inter-reader agreement, and reader confidence.

AB - OBJECTIVES: The purpose of this study was to establish whether an artificially intelligent (AI) system can be developed to automate stress echocardiography analysis and support clinician interpretation.BACKGROUND: Coronary artery disease is the leading global cause of mortality and morbidity and stress echocardiography remains one of the most commonly used diagnostic imaging tests.METHODS: An automated image processing pipeline was developed to extract novel geometric and kinematic features from stress echocardiograms collected as part of a large, United Kingdom-based prospective, multicenter, multivendor study. An ensemble machine learning classifier was trained, using the extracted features, to identify patients with severe coronary artery disease on invasive coronary angiography. The model was tested in an independent U.S.STUDY: How availability of an AI classification might impact clinical interpretation of stress echocardiograms was evaluated in a randomized crossover reader study.RESULTS: Acceptable classification accuracy for identification of patients with severe coronary artery disease in the training data set was achieved on cross-fold validation based on 31 unique geometric and kinematic features, with a specificity of 92.7% and a sensitivity of 84.4%. This accuracy was maintained in the independent validation data set. The use of the AI classification tool by clinicians increased inter-reader agreement and confidence as well as sensitivity for detection of disease by 10% to achieve an area under the receiver-operating characteristic curve of 0.93.CONCLUSIONS: Automated analysis of stress echocardiograms is possible using AI and provision of automated classifications to clinicians when reading stress echocardiograms could improve accuracy, inter-reader agreement, and reader confidence.

KW - Artificial Intelligence

KW - Coronary Artery Disease/diagnostic imaging

KW - Echocardiography/methods

KW - Humans

KW - Predictive Value of Tests

KW - Prospective Studies

U2 - 10.1016/j.jcmg.2021.10.013

DO - 10.1016/j.jcmg.2021.10.013

M3 - Article

C2 - 34922865

SN - 1936-878X

VL - 15

SP - 715

EP - 727

JO - JACC. Cardiovascular imaging

JF - JACC. Cardiovascular imaging

IS - 5

ER -

Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence

Abstract

Keywords

Access to Document

Fingerprint

Cite this