Deep convolutional neural networks identified eight common cardiac conditions from digital 12-lead ECGs with 92.9-100% accuracy and detected atrial fibrillation from ECG images with 96% accuracy.
Do deep convolutional neural networks accurately classify healthy and disease conditions from digital and image-based standard 12-lead ECGs?
Automated detection of multiple cardiac conditions using deep learning on standard digital or smartphone-imaged 12-lead ECGs is highly accurate and feasible.
Standard 12-lead electrocardiography (ECG) is used as the primary clinical tool to diagnose changes in heart function. The value of automated 12-lead ECG diagnostic approaches lies in their ability to screen the general population and to provide a second opinion for doctors. Yet, the clinical utility of automated ECG interpretations remains limited. We introduce a two-way approach to an automated cardiac disease identification system using standard digital or image 12-lead ECG recordings. Two different network architectures, one trained using digital signals (CNN-dig) and one trained using images (CNN-ima), were generated. An open-source dataset of 41,830 classified standard ECG recordings from patients and volunteers was generated. CNN-ima was trained to identify atrial fibrillation (AF) using 12-lead ECG digital signals and images that were also transformed to mimic mobile device camera-acquired ECG plot snapshots. CNN-dig accurately (92.9-100%) identified every possible combination of the eight most-common cardiac conditions. Both CNN-dig and CNN-ima accurately (98%) detected AF from standard 12-lead ECG digital signals and images, respectively. Similar classification accuracy was achieved with images containing smartphone camera acquisition artifacts. Automated detection of cardiac conditions in standard digital or image 12-lead ECG signals is feasible and may improve current diagnostic methods.
Глинер et al. (2020) studied Cardiac arrhythmias and morphological disorders (n=6,866). Deep convolutional neural networks (CNN-dig and CNN-ima) vs. Board-certified practicing cardiologists (ground truth) was evaluated on Classification accuracy for cardiac conditions. Deep convolutional neural networks identified eight common cardiac conditions from digital 12-lead ECGs with 92.9-100% accuracy and detected atrial fibrillation from ECG images with 96% accuracy.