Qualitative Simulation of Shock States in a Virtual Patient

In this paper, we describe the use of qualitative simulation to simulate shock states in a virtual patient. The system integrates AI techniques with a realistic visual simulation of the patient in a 3D environment representing an ER room. We have adapted qualitative process theory to the representation of physiological processes in order to be able to generate appropriate pathophysiological models. We describe how a subset of cardiac physiology can be modelled using qualitative process theory and discuss knowledge representation issues. We then present results obtained by the system and the benefits that can be derived from the use of a virtual patient in terms of training. Finally, we explore the problem of integrating multiple pathophysiological models for various aetiologies of shock states.