Non-linear fractal analysis of ECG signal collected from Maternal and normal sinus women

Abstract

Studies suggest that electrocardiogram (ECG) recordings are chaotic in nature and for years they are being studied using various non-linear methods. These approaches can be applied to study the nonlinear properties of ECG signals from pregnant women, termed as Maternal ECG (MECG). For a better perspective of how the nonlinear properties of MECG signal is different from normal ECG we include normal sinus ECG signal collected from a group of women. In this study we apply four different non-linear methods, namely Rescaled Range analysis (R/S), Higuchi's Fractal Dimension analysis (HFD), Detrended Fluctuation Analysis (DFA) and Multifractal Detrended Fluctuation Analysis (MFDFA) method to characterize non-linearity of the maternal ECG signal. Comparing the results of these methods we find that all the methods successfully distinguishes different properties of the ECG signals, but MF-DFA method being a multi-scale method, separates MECG signal from normal group most accurately.