Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network

Abstract

Computerized electrocardiogram (ECG) interpretation plays a critical role in the clinical ECG workflow¹. Widely available digital ECG data and the algorithmic paradigm of deep learning² present an opportunity to substantially improve the accuracy and scalability of automated ECG analysis. However, a comprehensive evaluation of an end-to-end deep learning approach for ECG analysis across a wide variety of diagnostic classes has not been previously reported. Here, we develop a deep neural network (DNN) to classify 12 rhythm classes using 91,232 single-lead ECGs from 53,549 patients who used a single-lead ambulatory ECG monitoring device. When validated against an independent test dataset annotated by a consensus committee of board-certified practicing cardiologists, the DNN achieved an average area under the receiver operating characteristic curve (ROC) of 0.97. The average F₁ score, which is the harmonic mean of the positive predictive value and sensitivity, for the DNN (0.837) exceeded that of average cardiologists (0.780). With specificity fixed at the average specificity achieved by cardiologists, the sensitivity of the DNN exceeded the average cardiologist sensitivity for all rhythm classes. These findings demonstrate that an end-to-end deep learning approach can classify a broad range of distinct arrhythmias from single-lead ECGs with high diagnostic performance similar to that of cardiologists. If confirmed in clinical settings, this approach could reduce the rate of misdiagnosed computerized ECG interpretations and improve the efficiency of expert human ECG interpretation by accurately triaging or prioritizing the most urgent conditions.

Extended Data Fig 1:

Deep Neural Network architecture. Our deep neural network consisted of 33 convolutional layers followed by a linear output layer into a softmax. The network accepts raw ECG data as input (sampled at 200 Hz, or 200 samples per second), and outputs a prediction of one out of 12 possible rhythm classes every 256 input samples.

Extended Data Fig 2:

Receiver operating characteristic curves for deep neural network predictions on 12 rhythm classes. Individual cardiologist performance is indicated by the red crosses and averaged cardiologist performance is indicated by the green dot. The line represents the ROC curve of model performance. AF-atrial fibrillation/atrial flutter; AVB-atrioventricular block; EAR-ectopic atrial rhythm; IVR-idioventricular rhythm; SVT-supraventricular tachycardia; VT-ventricular tachycardia. n = 7,544 where each of the 328 30-second ECGs received 23 sequence-level predictions.

Figure 1:

ROC and precision-recall curves. a, Examples of ROC curves calculated at the sequence level for atrial fibrillation (AF), trigeminy, and AVB. b, Examples of precision-recall curves calculated at the sequence level for atrial fibrillation, trigeminy, and AVB. Individual cardiologist performance is indicated by the red crosses and averaged cardiologist performance is indicated by the green dot. The line represents the ROC (a) or precision-recall curve (b) achieved by the DNN model. n = 7,544 where each of the 328 30-s ECGs received 23 sequence-level predictions.

Figure 2:

Confusion matrices. a, Confusion matrix for the predictions of the DNN versus the cardiology committee consensus. b, Confusion matrix for predictions of individual cardiologists versus the cardiology committee consensus. The percentage of all possible records in each category is displayed on a color gradient scale.