Artificial intelligence algorithm for detecting myocardial infarction using six-lead electrocardiography

Abstract

Rapid diagnosis of myocardial infarction (MI) using electrocardiography (ECG) is the cornerstone of effective treatment and prevention of mortality; however, conventional interpretation methods has low reliability for detecting MI and is difficulty to apply to limb 6-lead ECG based life type or wearable devices. We developed and validated a deep learning-based artificial intelligence algorithm (DLA) for detecting MI using 6-lead ECG. A total of 412,461 ECGs were used to develop a variational autoencoder (VAE) that reconstructed precordial 6-lead ECG using limb 6-lead ECG. Data from 9536, 1301, and 1768 ECGs of adult patients who underwent coronary angiography within 24 h from each ECG were used for development, internal and external validation, respectively. During internal and external validation, the area under the receiver operating characteristic curves of the DLA with VAE using a 6-lead ECG were 0.880 and 0.854, respectively, and the performances were preserved by the territory of the coronary lesion. Our DLA successfully detected MI using a 12-lead ECG or a 6-lead ECG. The results indicate that MI could be detected not only with a conventional 12 lead ECG but also with a life type 6-lead ECG device that employs our DLA.

Figure 1

Study flowchart. DLA deep learning-based algorithm, ECG electrocardiography, MI myocardial infarction.

Figure 2

Performances of artificial intelligence algorithms for detecting myocardial infarction. AUC area under the receiver operating characteristics curve, DLA deep learning-based algorithm, ECG electrocardiography, MI myocardial infarction, NPV negative predictive value, Sens sensitivity, Spec specificity, STEMI ST-segment elevation myocardial infarction, PPV positive predictive value, VAE variational autoencoder. These performances were calculated by using the operating point at Youden J statistics of development data.

Figure 3

Performances of artificial intelligence algorithms for detecting MI by coronary artery lesion. AE auto-encoder, AUC area under the receiver operating characteristics curve, DLA deep learning-based algorithm, ECG electrocardiography, LAD left anterior descending coronary artery, LCx left circumflex coronary artery, MI myocardial infarction, NPV negative predictive value, Sens sensitivity, Spec specificity, STEMI ST-segment elevation myocardial infarction, PPV positive predictive value, RCA right coronary artery. These performances were calculated by using the operating point at Youden J statistics of development data.

Figure 4

Reconstructed precordial 6-lead ECG using limb 6-lead ECG by auto-encoder and original precordial 6-lead ECG. DLA deep learning-based algorithm, ECG electrocardiography, MI myocardial infarction.

Figure 5

Sensitivity map of DLA for detecting MI. DLA deep learning-based algorithm, MI myocardial infarction.

Figure 6

Architecture of variational autoencoder and deep learning-based algorithm for detecting MI using a 12-lead or a 6-lead ECG. DLA deep learning-based algorithm, ECG electrocardiography, FC fully connected, MI myocardial infarction, VAE variational autoencoder.