WORK MODELING FOR CATEGORIZATION OF LATE POTENTIALS IN ECG SIGNALS BY WAY OF WAVELETS

Abstract

Late potentials in ECG take place in the terminal portion of the QRS complex and are characterized by tiny amplitudes and larger frequencies. The occurrence of late potentials may signify underlying distribution of electrical activity of the cells in the heart, and provides a substrate for production of arrhythmias. The conventional Fourier transform does not readily localize these features in time and frequency. Short Time Fourier Transform (STFT) is added useful because the concentration of the signal energy at various times in the cardiac cycle is more readily identified. This STFT suffers from the problem of selecting the proper window function as a window width can determine whether high temporal or high spectral resolution is achieved. The Wigner-Ville distribution, which produces a composite time – frequency distribution but suffers from the problem of interference from cross-terms. The problem of late potentials causes high levels of signal power to be seen at frequencies not representing the original signal. The present work describes the application of Wavelet Transform to provide a more accurate picture of the localized time-scale features indicative of the late potentials. The first step includes generating mathematical equations for various cases by developing a program in Matlab. Compared the signal under consideration with all those signals in the database by developing an identification code in Matlab. Analyzed the late potentials in the signal under consideration and identified the case. The second step includes generating mathematical equations for various specimen cases for the same type category by developing a program in Matlab. Compared the signal under consideration with all those signals in the database by developing an identification code in Matlab. Analyzed the late potentials in the signal under consideration and identified the case along with the specimen case.