Estimation of ApEn for short-term HFECG based on multi-resolution analysis

It is difficult to gain stable and credible values of ApEn (approximate entropy) for original signals of short-term high-frequency electrocardiogram (SHFECG) (X. Yinlin, N. Xinbao, C. Ying, W. Jun, Mode entropy and dynamical analysis of irregularity for HFECG. Chin. Sci. Bull. 49 (2004) 1886–1890, X. Ning, Y. Xu, J. Wang et al., Approximate entropy analysis of short-term HFECG based on wave mode. Physica A 346 (2005) 475–483). In this paper, original signals of SHFECG are decomposed into detail signals with multi-resolution wavelet analysis, and then their ApEn is calculated. A conclusion can be drawn that stable and credible values of ApEn can be counted from high-frequency detail signals (HFDS). Furthermore, different frequency details of SHFECG have different distributions of ApEn between heart diseases and healthy subjects with a variety of ages. Therefore, it is of great significance for diagnosing heart diseases in time.