Robustness evaluation of a dual-threshold QRS detection method for wearable ECG recorders

Wearable devices for real-time ECG monitoring and analysis require an accurate but simple QRS detection algorithm which does not consume too much computational load. Dual-threshold QRS detection techniques are one of the promising solutions to overcome this problem. This study investigates the performance and robustness of a dual-threshold QRS detection method under different levels of noise and physical activity. The algorithm was validated on synthetic noisy signal, available real ECG dataset, and ambulatory ECG records. The results demonstrate the potential of this low-computational algorithm for implementing in wearable devices if the SNR of noise is above 12 dB.