Correlation And Multifractal Analysis Of Motion Modulated HRV Signals

We can get the useful information and the conversion law between different states by the fractal analysis to the output signal of system.In order to study the nonlinear dynamical and other related characteristics in system conveniently. The system of Heart is one of the most complex systems in nature. It is responsible to provide the blood circulation in body long-lasting power. Research shows that, the complexity of the heart is result from the joint action by lots of self-regulatory organizations in the heart, and it is controled by sympathetic nerve and vagus nerve coordinatedly. This complex and regulate pattern can be represented by the variability of the output signal in the system of heart. This signal is named heart rate variability signal. This signal is a typical non-linear signal, and it is a chaotic signal that is overall ordered but partially disordered, which between deterministic signal and stochastic signal. Researching in heart rate variability signal can promote the application of non-linear analysis in medical signal field, and can find the relationship between the complexity of the heart and strength of autonomic nervous system. This research can provide valuable information for the theoretical study on cardiovascular disease. These years, with attention to physical and mental health, movement becomes an indispensiable part of the daily life. It has became a focus in this research field that how can ordinary people exercise to prevent cardiovascular diseases and how can athletes train to improve their activity levels.In this paper, we use the fractal method in nonlinear theory and practical experiment to establish the analyze model of HRV signal in motion modulation. We improved the fractal method and proposed the concept of HI-R/S analysis model and fractal coefficient by combining with the advantages of traditional R/S analysis method and Lo-R/S Analysis method. The new method solve the deficiencies that traditional methods can not quantify the short and long range correlation in HRV signals. We simulate the conversion law in cardiac autonomic nervous between different motions theoretically. At the same time, we combine the DFA and multifractal and discusses the multifractal characteristics in HRV signals and their change trends. We study the dynamic relationship among short-range correlation, long-range correlation, multifractal and autonomic regulation. The main conclusion as follow:(1). HI-R/S method include nonlinear dynamics information in the slope and interception of R/S curves and make up the deficiency that consider the Hurst index but ignore the information of interception in the traditional method. Comparing the result of HI-R/S, R/S method, and Lo-R/S method, we can see that the new method can better distinguish the fractal difference among HRV signal in different movement condition, and quantify the proportion of short-range correlation and long-range correlation in HRV signals. We draw the conclusion that the strength of the short-range correlation equal to the long-range correlation when the subjects before the motion and after the motion. The correlation in the motion is mainly represented by the long range correlation. This method is more suitable for describing the correlation of nonlinear signal in detail.(2). The general Hurst index calculated from DFA manifest that HRV signals have different intensity of multifractal properties under different motion states. Establish a bridge between correlation analysis and multi-fractal analysis by generalized Hurst exponent and Legendre transformation, and obtained the conclusions that multi-fractal characteristics in the system of heart during exercise is much more significant than that before and after exercise. Through comparing with result of the correlation analysis and the result of multifractal analysis, we can find that long-range correlation and multifractal have the same trend of change, but it is opposite to the trend of short-range correlation in HRV.(3). Through comparing with the result of correlation analysis and the result of multifractal analysis, we can find that the multifractality of each HRV signal in different states in afternoon are intenser than those in the morning. Combined with the meaning of long-range correlation and multifractal we can see that the regulation of the autonomic nervous in cardiac is strenger than that in the morning. In theory, nonlinear state of the heart in afternoon is more suitable for high-intensity exercise. Finally, we use the specific analytical experiments to prove that motion state and physiological rhythms has a certain degree of interaction effects to the complexity of the heart system.