Forecasting patient's behavior: a qualitative-fuzzy approach

by R. Bellazzi, L. Ironi, R. Guglielmann, M. Stefanelli



Proc. CESA98, IMACS-IEEE Multiconference


ABSTRACT

This paper describes the results of a novel approach for the identification of non-linear system dynamics when applied to the Blood Glucose metabolic system in insulin dependent diabetic patients. Such a method is based on fuzzy systems and qualitative models: the outcomes obtained from the simulation of a set of qualitative differential equations are used to automatically encode the available knowledge in a fuzzy rule-based system, which is then tuned to a set of experimental data. The results obtained show that the proposed framework can be reliably and efficiently used to forecast the Blood Glucose Level dynamics in response to different perturbations.



Back to publication list