Analyzing Time-Frequency Power Spectrum Limited in QRS Complex Based on the Wavelet Transform

It is well known that the high frequency components during QRS complex may relate with the scattered myocardial damages such as the focal necrosis. However, the time-frequency power spectrum limited in QRS can be hardly measured by use of the conventional fast Fourier transform (FFT) based algorithm. In contrast to the FFT, the wavelet transform (WT) has time-widths adapted to each frequency as window, which may help to solve the problem. Therefore we have developed software for analyzing the frequency profile of QRS, and with it we have tried to measure the latent myocardial damages. In this article, we also give clinical significance explanations with proposed method. The electrocardiogram (ECG) signals have been taken from some patients with intraventricular conduction abnormalities (IVCA), with or without myocardial infarction (MI) and non-Q-MI or Q-MI. The time dependence of the frequency powers spectrum is extracted by use of Morlet wavelet of 40 scales. The time-frequency power spectrum during QRS and the number of peaks of signals are calculated. The results have showed that the power spectrum at the low frequency ( ≤ 100Hz) power is reduced in IVCA, comparing with the normal cases. In IVCA with MI or MI, high frequency components were shown increased. As our conclusion, WT is beneficial for the analysis of masked frequency profile of QRS.