A decision support system for diabetes medicine selection using patient centered treatment based on fuzzy logic and domain ontology

Diabetes not only imposes psychological and physical pain on patients but also has high medical costs. Thus, the prescription strategy of clinical doctors must consider many factors. In this paper, we develop an individualized antidiabetic drugs recommendation system for patients with diabetes. This system combines fuzzy logic and an ontology system which can be manipulated with relative ease, and targets reasonable HbA1c levels that address individual differences among patients. The system was evaluated by an endocrinologist and an attending physician. That indicated the antidiabetic drugs recommendation system has good performance and is useful both in 90%. The system also performs 80% for accuracy, that can assist clinicians in the management of diabetes mellitus during selecting drugs and the patient individualization HbA1c Target.