The UNet-M-D model achieved up to 99.2% R-peak detection accuracy in noisy exercise ECGs and 77.4% accuracy in classifying exercise-induced fatigue using HRV features.
Does the UNet-M-D model improve R-peak detection and fatigue classification accuracy in noisy exercise ECGs?
The proposed UNet-M-D model provides highly accurate R-peak detection and fatigue classification in noisy exercise ECGs, offering a potential tool for sports health management.
Exercise-induced fatigue assessment via ECG classification relies on accurate R-peak detection for reliable HRV features. Addressing the lack of robust models for noisy exercise ECGs, we propose UNet-M-D, integrating positional encoding, multi-head self-attention, and dynamic convolution. Evaluated on GUDB and EPFL datasets, it achieves superior R-peak detection performance (up to 99.2% accuracy) with high noise resilience (6-18 SNR). Using optimally selected HRV features, our method attains 77.4% accuracy in fatigue classification, providing a scientific basis for sports health management and training adjustment.
Su et al. (2026) studied this question. The UNet-M-D model achieved up to 99.2% R-peak detection accuracy in noisy exercise ECGs and 77.4% accuracy in classifying exercise-induced fatigue using HRV features.