Approach for Cardiac Action Potential Detection from Noised Recordings

The morphology of action potential (AP) is a key element in the determination of normal or pathological processes in the cardiac cells and the diagnosis of heart diseases. Therefore obtaining useful information about the shape of AP pulses minimizing the noise contribution is an important task of signal processing to improve the diagnosis process. In this paper a method to detect the APs of cardiomyocytes is proposed, based on wavelet transform, decomposition and reconstruction in eigenvectors basis of cardiac cycles. The parallel conductance model is used to simulate the cardiac AP and ion currents taking into account white or pink noise. Numerical experiments demonstrate the performance of the AP detection for various signal-to-noise ratio (SNR). The proposed method can be used to obtain undistorted AP morphologies in order to interpret experimental studying of APs functional properties in cardiomyocytes by means of lab-on-chip technologies.