| Over the years, the Auto-Regressive Moving Average (ARMA) model based on linear assumption has established itself as the most popular technique for mathematical identification of the cardiovascular system. However, the cardiovascular system regulated by auto nervous functions is a complex, unclear, and non-linear system. In order to interpret and model it more correctly, non-linear and intelligent modeling techniques are needed into this research field.; In this study of cardiovascular dynamics, the identification of dynamic parameters involving the heart rate (HR), arterial blood pressure (ABP) and instantaneous lung volume (ILV), may be performed through a novel combination approach. This effective hybrid approach, which has parallel modular structure of Auto-regressive (AR) and neuro-fuzzy inference system (NFIS), becomes feasible to interpret cardiovascular linear mechanism of heart function and non-linear auto nervous regulation functions respectively. AR is an efficient method to identify the stationary time series function. Fuzzy set theory is very suitable to systems with uncertainties and expert knowledge. Neuro-fuzzy inference paradigm imports the auto-learning property into fuzzy logic engine, therefore extracts some hidden knowledge from data automatically.; The comparison between ARMA model and hybrid AR and NFIS model (ARNFIS), is investigated based on three subjects' clinical data in terms of accuracy, robustness, physiological meaning, and potential implementation in clinical problems. The AR-NFIS model has, in general, high accuracy, less sensitivity on patient physiological state, has the capability of modeling complex physiological system, and can be easily interpreted by physiological terminology in order to provide physiological meaningful results.; An application of this hybrid model, which involves a modified pulse wave contour cardiac output (CO) measurement, is also investigated. Based on AR-NFIS model simulation results, the parameters in pulse wave contour CO measurement are updated simultaneously to reach a higher-level accuracy in non-invasively continuous cardiac out measurement techniques. |
