Skip to Main Content
Boundary value detection in ECG signal is an invaluable part in ECG signal analysis for determining heart rate as well as various arrhythmias. This paper presents a novel approach of real time automated heart rate calculation through characteristic wave boundaries detection in ECG signal. For boundary detection, Empirical mode decomposition is used. Being a fully data driven adaptive technique, the present method depends on selection of proper and optimum set of IMFs to generate an intermediate signal. This study used empirical mode decomposition (EMD) for heart rate calculation and QRS complex duration in electrocardiogram signal. Before this analysis using EMD, to accomplish baseline correction and noise suppression with minimum signal distortion a modified morphological filtering (MMF) technique is used. The EMD-based technique gives results comparable to those obtained with the multiscale morphological derivative. In this technique EMD is performed on the MMF conditioned signal and the best intrinsic mode function (IMF) is chosen based on the cross-correlation between the noise removed signal and the IMFs. From the chosen IMF heart rate and QRS complex duration is calculated. We conclude that the EMD-based algorithm is a preferred technique considering algorithm simplicity, robustness, detection accuracy.