Model identification of the neural control of the cardiovascular system using NARMAX models

The neural control of the human cardiovascular system has been modeled using a NARMAX (non-linear autoregressive moving-average model with exogenous inputs) model, and compared with MA (moving-average) model and an ARMAX (autoregressive moving-average model with exogenous inputs) model. In these approaches, models are sought in the form of difference equations with unknown parameters to be estimated on the basis of input-output data. The input variable considered has been the carotid-sinus blood pressure and five output variables have been analyzed: heart rate control, peripheral resistance control, myocardial contractility control, venous tone control, and coronary resistance control. Results obtained show that NARMAX models present a better fit than the ARMAX models