Building an enhanced case-based reasoning and rule-based systems for medical diagnosis

Abstract

Expert systems are computer programs that use knowledge and reasoning to solve problems typically solved by human experts. Expert systems have been used in medicine to diagnose diseases, recommend treatments, and plan surgeries. Interpretability of the results in medical applications is crucial since the decision that will be taken based on the system’s output has a direct effect on people’s health and lives which makes expert systems ideal choices when dealing with these applications in contrast to other machine learning approaches. An expert system has the ability to explain its own line of reasoning providing a robust way of diagnosis. This paper presents two types of expert systems for medical diagnosis. The first system is a case-based reasoning system using a database of previously diagnosed cases to diagnose a new case. The second system is a rule-based expert system that uses a set of if–then rules extracted from a decision tree classifier to make diagnoses. In this paper, machine learning-based similarity functions are proposed and compared with other traditional similarity functions. The results of this study suggest that expert systems can be a valuable tool for medical diagnosis. The two systems presented in this paper achieved competitive results, and they provide diagnoses similar to those made by human experts.