Parameter estimation of systemic vascular bed by arterial-venous pressure transfer function

Purpose of this paper was to obtain a low-order identification-oriented model of the systemic vascular bed, with a future view to beat-to-beat monitoring and control applications. To this aim reduced models of the systemic bed were compared on the basis of Akaike Information Criterion (AIC), combined with parameter estimation accuracy and physical interpretability. These criteria have been applied to systemic pressure data generated by a cardiovascular simulator, implemented using a Computer Aided Design of Control System (CADCS) package. The results obtained show that only simple models can be accurately estimated, when signals with a limited frequency spectrum such as the physiological ones are used.