Nonlinear Analysis Of HRV Signal Based On The Generalized Complexity Measure

ObjectivesThe heart rate variability(HRV) reflects subtle time changes from beat to beat and represents the fluctuant characteristic and the periodic variation of the sinus heart rate during a certain period of time.These fluctuations of the heart rate don't happen at random,but are controlled by the body nerves and body fluid,which are the responses for being adapted to different physiological conditions or some pathological conditions. The fluctuations of the heart rate are influenced by many factors such as high-level neural activity,the spontaneous activity of central nervous system,respiratory activities, as well as the cardiovascular reverse activities affected by the pressure and chemoreceptor,and these factors impose the synthetical function on the SNS and the PNS to form heart rate variability.In other words,the HRV contains a lot of information and can be measured and analyzed quantificationally to evaluate the tension and balance of the parasympathetic-sympathetic nerve and to characterize the activity of the heart and blood system.Therefore,HRV is a valuable index to evaluate the heart and blood system and to predict cardiac sudden death and arrhythmia.It has been wildly confirmed and accepted that the heart and blood system possesses significant nonlinear dynamic characteristics and the HRV signal contains the time-series output from the heart and blood system and are related to regulation model of the heart and blood system and a number of parameters.Both time-domain and frequency-domain analysis of HRV reflect the overall changes in heart rate and nonlinearity analysis can reveal more information contained in ECG and reflect the complexity from beat to beat and represent the instantaneous changes in heart rate.So HRV can be described by parameters in nonlinear systems.Considering the HRV signal contains such special features as the large dynamic range,short steady-state residence time and the short data length,in a number of non-linear dynamic parameters, the complexity measure is frequently used to characterize the HRV signal.Complexity measures in the representative of traditional Kc aim at the randomicity of sequence,but not the complexity.As the above stated,this study focuses on a novel complexity measure-Generalized Complexity Measure(GC) and applies it to the analysis of HRV. Calculation and comparison of GC between the normal group and the PB group may set up a non-invasive quantitative index for diagnosing cardiovascular disease and improve the understanding of the underlying physiological mechanism of the HRV. Furthermore,introduction of the GC will provide a theoretical foundation for clinical ECG medical diagnosis and medication.MethodsThe study object of HRV is the changes and the regularity of beat to beat,and a list of the differences during every cardiac cycle can present a lot of parameters out of order and reflect the continuous instantaneous fluctuations of heart rate.How to obtain and analyze the regularity from the fluctuations of the heart rate and these disorder parameters to illustrate the body's physiological or pathological changes is the essence of the HRV analysis and is also an important point different from other ECG technology.HRV measurement is actually measurement of cardiac cycle variation,and the measurement of the HRV indexes are based on the measurement of R-R intervals.1.Extraction and group of HRV signal;2.Coarse-grain numerical sequence of R-R intervals into symbolic sequence;3.Program with Visual C++ 6.0;4.Calculation of GC of the two sample groups;5.Find the regularity and establish quantitative indexes;6.Statistical analysis.Results1.It presents ECG modality of two sample groups and the corresponding distribution of the numerical sequence of R-R intervals;2.The statistical analysis of GC of the two sample groups shows: GC_T=13.0799±1.5768(PB),GC_T=37.2474±0.5525(Normal),There is significant difference(P＜0.01);3.The picture of scatter plot shows that there is a big dedifferentiation between GC_R of the two sample groups,and rough distinguish between GC_C of the normal sample and the PB sample,with overlap of only some data points,while there is clear distinguish between GC_T of the two sample groups.ConclusionThis research shows that Transferability Complexity(GC_T),as one of GC,can distinguish between the HRV of normal samples and the PB samples and quantificationally characterize the complexity of structure in HRV signal.The broad foreground and potential value is reviewed to analyze the heart rate variability on the basis of non-linear dynamics and complexity science.It's expected to become a non-invasive objective index for diagnosing cardiovascular disease.