Abstract
Carotid intima-medial thickness (CIMT) allows for early detection of carotid atherosclerotic vascular disease. Using fuzzy logic techniques, we propose a model for calculating the risk for predicting CIMT using expert opinion and various factors associated with CIMT in recently menopausal women. Study participants were enrolled in the Kronos Early Estrogen Prevention Study (KEEPS), a four year study where participants were assigned to either oral or transdermal hormone therapy or to the placebo group. The fuzzy mathematical model uses five fuzzy logic methods to calculate this risk factor for CIMT, including the analytic hierarchy process (AHP) method, Guiasu method, Yen method, Dempster-Shafer (DS) method, and the set valued statistical method (SVSM). The four methods, AHP, Guiasu, Yen, DS, and SVSM, were correlated to each other according to the Pearson ranking method. In addition, results showed that the estrogen treatment groups showed a statically significant decrease in calculated CIMT risk compared to placebo when all the methods were combined together (p = 0.0017). Our fuzzy mathematical model provides the first fuzzy mathematical model to predict the risk of CIMT in recently menopausal women and menopausal hormone therapy may be beneficial in reducing this risk.