A Bond Graph Approach to the Modelling of Cardiovascular System with Embedded Autonomic Nervous System

Abstract

The study of control of cardiovascular system (CVS) by autonomic nervous system (ANS) has been very useful in detection of different types of cardiovascular disorders. The data gathered during different tests of the ANS like Valsalva manoeuvre and tilt test etc. are not so easy to analyse, mainly due to the very complex mechanisms involved in the regulation of the CVS. However, model-based analysis of test obtained data has been proposed in a few literature in order to cope with this problem but only a very few models have successfully demonstrated the exact anatomical working of the CVS. In this thesis, a very basic model of CVS is presented with the dynamics of blood flowing throughout the body. A new model of the ANS, representing the baroreceptors, the sympathetic action and ventricular muscles is also presented. The models have been developed using the bond graph formalism, as it offers a simple and unified representation for all energy domains and facilitates the integration of mechanical, hydraulic and electrical phenomenon. ANS regulation of cardiac contractility is represented by means of continuous transfer functions. The results in terms of cardiac contractility and end systolic pressure (ESP) are presented for two different experiments. Another model is proposed to represent ANS action in terms of an overwhelming controller. Such a controller can also be used in devices like left ventricular assist device.