Medical expert system based ontology for diabetes disease diagnosis

Diabetes has recently played an important role in the considerable indisposition of humanity. The diabetes patients are in a condition resulting from defects in insulin secretion with long range harm, dysfunction and failure of various human organs. Potential medical diagnosis of the disease diabetes is essential, including medical applications. However, limitations in representing the meaning and content of the domain knowledge for support of the search process for knowledge from a variety of user perceptions are considerable among these applications. In this paper, a medical expert system for diagnosis of diabetes is proposed. The diabetes ontology is developed using the OWL format with 9 sub-classes. The interval results with the weighted OWA similarity algorithm are expressed for easy interpretation by users. The expert system is developed in the form of a web-based application with web service architecture. An overall consistency rate of 90.7% was achieved with test data from 65 patients. The obtained results properly show that the system can help to diagnose diabetes early on and serve as a guide for people with diabetes to monitor the disease. It is quite possible that the proposed medical expert system discussed in the current work can be properly applied to other biological datasets to deal with the problems of disease diagnosis effectively.