Web-based CBR System for Support Medical Diagnosis

Abstract

From the early days of development of Artificial Intelligence, there was a strong in-terest in applications in the area of medicine. The interest was strong enough to form a separate branch in the early ‘80s entitled Artificial Intelligence in Medicine (AIM). DXplain [1] is an illustration of system from this early period. It is an internal medi-cine expert system developed at the Massachusetts General Hospital that is still in use at a number of hospitals and medical schools, mostly for clinical education purposes. Rather than using an expert system approach or another rule-inferring paradigm we decided to employ a Case-based Reasoning (CBR) methodology [4]. Storing and searching among past cases has an advantage in complex domains where it is difficult to create a global theory that explains most of the existing cases.