A new real-time R-wave detection algorithm based on integral projection function

Abstract

A new algorithm for detection of the R-wave of electrocardiogram (ECG) is proposed in this paper. It includes three important steps: data normalization, filter and detection. Data normalization makes the algorithm can be used on fixed-point instruments and reduces the computational complexity. Cohen Daubechies Feauveau 9/7(CDF9/7) Wavelet Filter, an integer wavelet filter, reduces false detection caused by the various types of interference present in ECG signal. In the detection of R-wave, to set the threshold self-adaptively, a new method based on the integral projection function of extreme points is proposed. In addition, a method of ECG segmentation is chosen to simplify the process in the new algorithm. For the standard MIT/BIH arrhythmia database, this new algorithm correctly detects 98.89 percent of the R-wave. Comparing with other algorithms, implementation of the algorithm is significantly simplified while the detection accuracy is favorable. The new algorithm is more suitable for the real-time processing in portable ECG instruments and it will not lose the important information of the original ECG signal.