A 34-layer convolutional neural network exceeded the average performance of 6 board-certified cardiologists in both recall and precision for detecting arrhythmias from single-lead wearable ECGs.
Does a 34-layer convolutional neural network improve arrhythmia detection from single-lead wearable monitor ECGs compared to individual board-certified cardiologists?
A deep learning algorithm can detect a wide range of arrhythmias from single-lead ECGs with higher sensitivity and precision than average board-certified cardiologists.
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).
Rajpurkar et al. (Thu,) conducted a other in Heart arrhythmias. 34-layer convolutional neural network vs. 6 individual board-certified cardiologists was evaluated on Recall (sensitivity) and precision (positive predictive value) for detecting heart arrhythmias. A 34-layer convolutional neural network exceeded the average performance of 6 board-certified cardiologists in both recall and precision for detecting arrhythmias from single-lead wearable ECGs.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: