A Novel Dual-Slope Based Electrocardiogram Peak Detection Method for Wearable Devices

Abstract

This paper presents a new electrocardiogram (ECG) peak detection method based on Dual-Slope processing with low computational complexity, suitable for wearable devices. The method calculates the slopes on both sides of a data point, and only peak points can have a large response to this processing. The sliding window integral and low pass filter method are used to make the signal smoother and more suitable for peak detection. Two adaptive thresholds are used to justify the peak point. The algorithm, evaluated by using data in MIT/BIH Database, achieves a detection rate of 99.49%, a sensitivity of 99.74%.