A Fuzzy Logic System with Attribute Ranking Technique for Risk-Level Classification of CAHD in Female Diabetic Patients

Complex problems in various disciplines like biology, medicine, humanities, management studies and so on gives various research dimensions in soft computing. Risk classification is one of the thrust area in the field of medicine. This research work aims in risk classification of coronary artery heart disease in female diabetic patients using fuzzy logic. Fuzzy logic which is a component of soft computing is used for classification. Female patients those are having diabetes mellitus (DM) have a high occurrence of coronary artery heart disease (CAHD), since diabetes is implicated in the formation of atherosclerotic plaque. The input parameters are plasma glucose concentration, diastolic blood pressure, body mass index and age are taken as input parameters for designing Mamdani type fuzzy inference system. 25 numbers of rules are given for the risk prediction. The risk of CAHD is predicted as low, intermediate and high. The implementation is carried out through MATLAB 2012a. The PIMA women diabetes dataset is taken for simulation. The performance of the proposed risk classifier is measured in terms of classification accuracy, sensitivity and specificity. Also the outputs are demonstrated by rule viewer and surface viewer.