Design and implementation of a high quality R-peak detection algorithm

Abstract

In modern medicine, electrocardiogram (ECG) is an important way to diagnose cardiovascular disease and monitor health information. The detection of R-peak is very important in ECG signal processing. To improve the accuracy and sensitivity of detection, a compound algorithm with high quality is presented in this paper. The algorithm removes high frequency noise and power frequency noise through an IIR low-pass filter, then do wavelet transform to the filtered signal. Adaptive threshold was used to extract modulus maxima. Rechecking is applied when there are mistakes. Additionally, template matching method is exploited in the rechecking to false detection. The algorithm is evaluated by using MIT-BIH arrhythmia database [1]. Finally, we obtained sensitivity of 99.79% and accuracy of 99.81%.