Screening for cardiac contractile dysfunction using an artificial intelligence–enabled electrocardiogram

Zachi I. Attia, Suraj Kapa, Francisco Lopez-Jimenez, Paul McKie, Dorothy J. Ladewig, Gaurav Satam, Patricia Pellikka, Maurice E Sarano, Peter Noseworthy, Thomas M. Munger, Samuel J Asirvatham, Christopher G. Scott, Rickey E. Carter, Paul Andrew Friedman

Research output: Contribution to journalLetter

34 Citations (Scopus)

Abstract

Asymptomatic left ventricular dysfunction (ALVD) is present in 3–6% of the general population, is associated with reduced quality of life and longevity, and is treatable when found1–4. An inexpensive, noninvasive screening tool for ALVD in the doctor’s office is not available. We tested the hypothesis that application of artificial intelligence (AI) to the electrocardiogram (ECG), a routine method of measuring the heart’s electrical activity, could identify ALVD. Using paired 12-lead ECG and echocardiogram data, including the left ventricular ejection fraction (a measure of contractile function), from 44,959 patients at the Mayo Clinic, we trained a convolutional neural network to identify patients with ventricular dysfunction, defined as ejection fraction ≤35%, using the ECG data alone. When tested on an independent set of 52,870 patients, the network model yielded values for the area under the curve, sensitivity, specificity, and accuracy of 0.93, 86.3%, 85.7%, and 85.7%, respectively. In patients without ventricular dysfunction, those with a positive AI screen were at 4 times the risk (hazard ratio, 4.1; 95% confidence interval, 3.3 to 5.0) of developing future ventricular dysfunction compared with those with a negative screen. Application of AI to the ECG—a ubiquitous, low-cost test—permits the ECG to serve as a powerful screening tool in asymptomatic individuals to identify ALVD.

Original languageEnglish (US)
Pages (from-to)70-74
Number of pages5
JournalNature Medicine
Volume25
Issue number1
DOIs
StatePublished - Jan 1 2019

Fingerprint

Left Ventricular Dysfunction
Electrocardiography
Ventricular Dysfunction
Artificial Intelligence
Screening
Artificial intelligence
Stroke Volume
Area Under Curve
Hazards
Odds Ratio
Quality of Life
Confidence Intervals
Neural networks
Costs and Cost Analysis
Sensitivity and Specificity
Population
Costs

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)

Cite this

@article{f50b9478709f4404a3fa93cf60fba2a7,
title = "Screening for cardiac contractile dysfunction using an artificial intelligence–enabled electrocardiogram",
abstract = "Asymptomatic left ventricular dysfunction (ALVD) is present in 3–6{\%} of the general population, is associated with reduced quality of life and longevity, and is treatable when found1–4. An inexpensive, noninvasive screening tool for ALVD in the doctor’s office is not available. We tested the hypothesis that application of artificial intelligence (AI) to the electrocardiogram (ECG), a routine method of measuring the heart’s electrical activity, could identify ALVD. Using paired 12-lead ECG and echocardiogram data, including the left ventricular ejection fraction (a measure of contractile function), from 44,959 patients at the Mayo Clinic, we trained a convolutional neural network to identify patients with ventricular dysfunction, defined as ejection fraction ≤35{\%}, using the ECG data alone. When tested on an independent set of 52,870 patients, the network model yielded values for the area under the curve, sensitivity, specificity, and accuracy of 0.93, 86.3{\%}, 85.7{\%}, and 85.7{\%}, respectively. In patients without ventricular dysfunction, those with a positive AI screen were at 4 times the risk (hazard ratio, 4.1; 95{\%} confidence interval, 3.3 to 5.0) of developing future ventricular dysfunction compared with those with a negative screen. Application of AI to the ECG—a ubiquitous, low-cost test—permits the ECG to serve as a powerful screening tool in asymptomatic individuals to identify ALVD.",
author = "Attia, {Zachi I.} and Suraj Kapa and Francisco Lopez-Jimenez and Paul McKie and Ladewig, {Dorothy J.} and Gaurav Satam and Patricia Pellikka and Sarano, {Maurice E} and Peter Noseworthy and Munger, {Thomas M.} and Asirvatham, {Samuel J} and Scott, {Christopher G.} and Carter, {Rickey E.} and Friedman, {Paul Andrew}",
year = "2019",
month = "1",
day = "1",
doi = "10.1038/s41591-018-0240-2",
language = "English (US)",
volume = "25",
pages = "70--74",
journal = "Nature Medicine",
issn = "1078-8956",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Screening for cardiac contractile dysfunction using an artificial intelligence–enabled electrocardiogram

AU - Attia, Zachi I.

AU - Kapa, Suraj

AU - Lopez-Jimenez, Francisco

AU - McKie, Paul

AU - Ladewig, Dorothy J.

AU - Satam, Gaurav

AU - Pellikka, Patricia

AU - Sarano, Maurice E

AU - Noseworthy, Peter

AU - Munger, Thomas M.

AU - Asirvatham, Samuel J

AU - Scott, Christopher G.

AU - Carter, Rickey E.

AU - Friedman, Paul Andrew

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Asymptomatic left ventricular dysfunction (ALVD) is present in 3–6% of the general population, is associated with reduced quality of life and longevity, and is treatable when found1–4. An inexpensive, noninvasive screening tool for ALVD in the doctor’s office is not available. We tested the hypothesis that application of artificial intelligence (AI) to the electrocardiogram (ECG), a routine method of measuring the heart’s electrical activity, could identify ALVD. Using paired 12-lead ECG and echocardiogram data, including the left ventricular ejection fraction (a measure of contractile function), from 44,959 patients at the Mayo Clinic, we trained a convolutional neural network to identify patients with ventricular dysfunction, defined as ejection fraction ≤35%, using the ECG data alone. When tested on an independent set of 52,870 patients, the network model yielded values for the area under the curve, sensitivity, specificity, and accuracy of 0.93, 86.3%, 85.7%, and 85.7%, respectively. In patients without ventricular dysfunction, those with a positive AI screen were at 4 times the risk (hazard ratio, 4.1; 95% confidence interval, 3.3 to 5.0) of developing future ventricular dysfunction compared with those with a negative screen. Application of AI to the ECG—a ubiquitous, low-cost test—permits the ECG to serve as a powerful screening tool in asymptomatic individuals to identify ALVD.

AB - Asymptomatic left ventricular dysfunction (ALVD) is present in 3–6% of the general population, is associated with reduced quality of life and longevity, and is treatable when found1–4. An inexpensive, noninvasive screening tool for ALVD in the doctor’s office is not available. We tested the hypothesis that application of artificial intelligence (AI) to the electrocardiogram (ECG), a routine method of measuring the heart’s electrical activity, could identify ALVD. Using paired 12-lead ECG and echocardiogram data, including the left ventricular ejection fraction (a measure of contractile function), from 44,959 patients at the Mayo Clinic, we trained a convolutional neural network to identify patients with ventricular dysfunction, defined as ejection fraction ≤35%, using the ECG data alone. When tested on an independent set of 52,870 patients, the network model yielded values for the area under the curve, sensitivity, specificity, and accuracy of 0.93, 86.3%, 85.7%, and 85.7%, respectively. In patients without ventricular dysfunction, those with a positive AI screen were at 4 times the risk (hazard ratio, 4.1; 95% confidence interval, 3.3 to 5.0) of developing future ventricular dysfunction compared with those with a negative screen. Application of AI to the ECG—a ubiquitous, low-cost test—permits the ECG to serve as a powerful screening tool in asymptomatic individuals to identify ALVD.

UR - http://www.scopus.com/inward/record.url?scp=85059768809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059768809&partnerID=8YFLogxK

U2 - 10.1038/s41591-018-0240-2

DO - 10.1038/s41591-018-0240-2

M3 - Letter

C2 - 30617318

AN - SCOPUS:85059768809

VL - 25

SP - 70

EP - 74

JO - Nature Medicine

JF - Nature Medicine

SN - 1078-8956

IS - 1

ER -