Deep CNN-based detection of cardiac rhythm disorders using PPG signals from wearable devices

Summary

This PLOS ONE paper presents a deep 1D-CNN for multi-class cardiac rhythm classification from wrist PPG signals, targeting practical wearable device deployment. Beyond binary AFib detection, the model simultaneously classifies premature atrial contractions (PAC) and premature ventricular contractions (PVC) — a more clinically comprehensive output than most competing systems. The model is evaluated on the UCI ML PPG dataset combined with additional wearable device recordings and achieves sensitivity ~96% and specificity ~95% for AFib detection.