A Review on the Nonlinear Dynamical System Analysis of Electrocardiogram Signal

Abstract

Electrocardiogram (ECG) signal analysis has received special attention of the researchers in the recent past because of its ability to divulge crucial information about the electrophysiology of the heart and the autonomic nervous system activity in a noninvasive manner. Analysis of the ECG signals has been explored using both linear and nonlinear methods. However, the nonlinear methods of ECG signal analysis are gaining popularity because of their robustness in feature extraction and classification. The current study presents a review of the nonlinear signal analysis methods, namely, reconstructed phase space analysis, Lyapunov exponents, correlation dimension, detrended fluctuation analysis (DFA), recurrence plot, Poincaré plot, approximate entropy, and sample entropy along with their recent applications in the ECG signal analysis.

Figure 1

Various types of application of nonlinear dynamical system analysis of ECG.

Figure 2

A representative RRI time series obtained from a 5 min ECG signal.

Figure 3

Computation of the optimal embedding dimension by the method of false nearest neighbours. The optimal embedding dimension was 7, and the corresponding percent false neighbour was 44.83%. The method of false nearest neighbour was implemented using Visual Recurrence Analysis freeware (V4.9, USA), developed by Kononov [37].

Figure 4

Optimal time delay computation by the first minimum of the AMIF. The first minimum of the AMIF was 2. The AMIF was calculated using Visual Recurrence Analysis freeware (V4.9, USA), developed by Kononov [37].

Figure 5

3D phase space attractor of an RRI time series. The attractor was plotted using the MATLAB Toolbox developed by Yang [39].

Figure 6

Illustration of the first 3 stages during the construction of a Cantor set in 2D: (a) n = 0, (b) n = 1, and (c) n = 2 [48].

Figure 7

Correlation dimensions of the reconstructed phase space plot of RRI time series at different embedding dimensions. The correlation dimensions were calculated using Visual Recurrence Analysis freeware (V4.9, USA), developed by Kononov [37].

Figure 8

Log-log graph of F(n) versus n for RRI time series. The graph was plotted using Biomedical Workbench toolkit of LabVIEW (National Instruments, USA).

Figure 9

Recurrence plot of an RRI time series. The recurrence plot was generated using the MATLAB Toolbox developed by Yang [39].

Figure 10

The Poincaré plot of the RRI time series represented in Figure 2. The plot was generated using Biomedical Workbench toolkit of LabVIEW (National Instruments, USA).