The Deterministic Chaos in Heart Rate Variability Signal and Analysis Techniques

The heart rate variability (HRV) signal carries important information about the systems controlling the heat rate, mainly elicited by autonomic nervous system (sympathetic and parasympathetic) controls. The analysis of HRV signal with existing linear techniques based on fast Fourier transform and autoregressive model have the disadvantages of lower sensitivity and positive predictive value of <30%. These may be due to methodological errors, such as assumption of RR-interval series as stationary, re-sampling of RR-interval series, thereby adding errors into the resulting analysis. The R-R interval series which formulates the heart rate variability signal is actually a nonstationary signal. To enhance the accuracy of cardiac health prognosis, the non-linear methods are currently being proposed for the HRV signal analysis, which uses the actual RR-interval series. This paper reviews the non-linear methods of HRV analysis such as correlation dimension, largest Lyupnov exponent, power law slope, fractal analysis, detrended fluctuation analysis, complexity measure etc. The drawback is these methods are that, long term electrocardiogram recording is required for efficient heart health prognosis with these methods. General Terms Review on cardiac variability studies.