Motion-Robust Multimodal Fusion of PPG and Accelerometer Signals for Three-Class Heart Rhythm Classification
This work addresses the challenge of robust heart rhythm monitoring for elderly patients with atrial fibrillation, offering incremental improvements in classification accuracy in noisy, real-world conditions.
The paper tackled the problem of classifying heart rhythms using wrist-worn PPG signals, which are prone to motion artifacts, by introducing RhythmiNet, a neural network that fuses PPG and accelerometer data with attention modules, achieving a 4.3% improvement in macro-AUC over a PPG-only baseline and a 12% improvement over a handcrafted feature model.
Atrial fibrillation (AF) is a leading cause of stroke and mortality, particularly in elderly patients. Wrist-worn photoplethysmography (PPG) enables non-invasive, continuous rhythm monitoring, yet suffers from significant vulnerability to motion artifacts and physiological noise. Many existing approaches rely solely on single-channel PPG and are limited to binary AF detection, often failing to capture the broader range of arrhythmias encountered in clinical settings. We introduce RhythmiNet, a residual neural network enhanced with temporal and channel attention modules that jointly leverage PPG and accelerometer (ACC) signals. The model performs three-class rhythm classification: AF, sinus rhythm (SR), and Other. To assess robustness across varying movement conditions, test data are stratified by accelerometer-based motion intensity percentiles without excluding any segments. RhythmiNet achieved a 4.3% improvement in macro-AUC over the PPG-only baseline. In addition, performance surpassed a logistic regression model based on handcrafted HRV features by 12%, highlighting the benefit of multimodal fusion and attention-based learning in noisy, real-world clinical data.