Detection of QRS complex in electrocardiogram signal by the empirical mode decomposition

The acquisition and the pretreatment of the signals (physiological signals) are often followed by the extraction of the parameters of clinical importance. In the case of the electrocardiogram signal (ECG), the QRS complex is one of the most significant parameters for the diagnosis of the cardiac arrhythmias. In this paper the method known as the empirical mode decomposition (EMD) is used for detection of QRS complex. This method aims to decompose nonlinear and non stationary signals adaptively in a series of signals modulated in amplitude and frequency called intrinsic mode function (IMF). We tested our algorithm on several signals of the MIT/BIH database comprising different pathologies, we obtained a sensitivity of 78% to 99%.