Cardiologist-Level Arrhythmia Detection with Convolutional Neural Networks

TL;DR

This paper presents a deep learning algorithm that surpasses cardiologists in detecting arrhythmias from ECG data, using a large dataset and a 34-layer CNN to achieve superior accuracy.

Contribution

The study introduces a novel CNN-based method trained on an extensive ECG dataset, outperforming expert cardiologists in arrhythmia detection.

Findings

Outperforms cardiologists in recall and precision

Uses a large, diverse ECG dataset

Employs a 34-layer convolutional neural network

Abstract

We develop an algorithm which exceeds the performance of board certified cardiologists in detecting a wide range of heart arrhythmias from electrocardiograms recorded with a single-lead wearable monitor. We build a dataset with more than 500 times the number of unique patients than previously studied corpora. On this dataset, we train a 34-layer convolutional neural network which maps a sequence of ECG samples to a sequence of rhythm classes. Committees of board-certified cardiologists annotate a gold standard test set on which we compare the performance of our model to that of 6 other individual cardiologists. We exceed the average cardiologist performance in both recall (sensitivity) and precision (positive predictive value).