Despite the use of Electronic Medical Record Systems (EMRs) in primary care, physicians
still lack decision support tools to help with decision making in the delivery of health care. In this paper we propose a framework for an Intelligent Decision Support system that uses
hybrid architecture and combines the concepts of data mining of knowledge bases (KB) and artificial neural networks (ANN). The model is presented in the context of the primary health care system, with an aim to create and track patient profiles for use in pattern recognition to identify unusual test readings and trigger alerts, support decision making by recalling past information, produce domain knowledge from the recalled information, perform reasoning from “new” domain knowledge and serve as a predictive tool in decision support.
Our approach focuses first on building descriptive and predictive models for the particular
domain, and then using these models to formulate the hybrid system. We present a case
study to show how the system would be applied in a clinical setting.