Key result
A deep convolutional neural network trained on over 2 million 12-lead ECGs achieved F1 scores above 80% and specificity over 99% for detecting six types of abnormalities, outperforming 4th-year cardiology residents in 5 out of 6 classes.
Why the study?
Short-duration 12-lead ECGs provide a full evaluation of heart activity but had not been studied in previous end-to-end machine learning papers.
Does a deep convolutional network improve the diagnosis of 6 ECG abnormalities in short-duration 12-lead ECGs compared to cardiology residents?
Does a deep convolutional network improve the diagnosis of 6 ECG abnormalities in short-duration 12-lead ECGs compared to cardiology residents?
An end-to-end deep learning model trained on over 2 million 12-lead ECGs can accurately detect 6 common abnormalities, outperforming cardiology residents.
We present a model for predicting electrocardiogram (ECG) abnormalities in short-duration 12-lead ECG signals which outperformed medical doctors on the 4th year of their cardiology residency. Such exams can provide a full evaluation of heart activity and have not been studied in previous end-to-end machine learning papers. Using the database of a large telehealth network, we built a novel dataset with more than 2 million ECG tracings, orders of magnitude larger than those used in previous studies. Moreover, our dataset is more realistic, as it consist of 12-lead ECGs recorded during standard in-clinics exams. Using this data, we trained a residual neural network with 9 convolutional layers to map 7 to 10 second ECG signals to 6 classes of ECG abnormalities. Future work should extend these results to cover a large range of ECG abnormalities, which could improve the accessibility of this diagnostic tool and avoid wrong diagnosis from medical doctors.
No takes yet. Share an insight, caveat, or question.
Ribeiro et al. (2018) studied ECG abnormalities (n=1,676,384). Deep Convolutional Neural Network vs. 4th year cardiology resident medical doctors was evaluated on Diagnostic performance (F1 score, precision, recall, specificity) for 6 ECG abnormalities. A deep convolutional neural network trained on over 2 million 12-lead ECGs achieved F1 scores above 80% and specificity over 99% for detecting six types of abnormalities, outperforming 4th-year cardiology residents in 5 out of 6 classes.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: