APPLICATION OF FUZZY KNOWLEDGE GRAPH IN DIAGNOSIS SUPPORT FOR DIABETES PATIENTS

Abstract

Recently, the needs of human in health care are extremely necessary. Among common diseases, Diabetes is one of the dangerous diseases and many people are suffering from this disease. Age, obesity, lack of exercise, hereditary diabetes, lifestyle, unreasonable diet, high blood pressure, etc. are all causes of Diabetes. People, who have diabetes, are at high risk of developing diseases like heart disease, kidney disease, stroke, eye problems, nerve damage, etc. Information technology tools used in diagnostic support help doctors detect a patient's disease condition quickly and accurately. Based on the diagnosis, the appropriate treatment regimens for patients can be quickly indicated. In this article, we focus on researching fuzzy knowledge graph models in supporting diabetes diagnosis. The knowledge graph model, mainly based on graph theory combined with fuzzy inference, is a new research direction in recent years. This model is applied to detect potentially data patterns that are prone to disease and assist doctors in diagnosis. To evaluate the performance of the proposed model, the implementations are performed on a data set collected from the doctors at Hung Yen General Hospital. Experimental results show that our new model gives better results than the compared models.