Identification of the Framingham Risk Score by an Entropy-Based Rule Model for Cardiovascular Disease.

摘要: Since 2001, cardiovascular disease (CVD) has had the second-highest mortality rate, about 15,700 people per year, in Taiwan. It thus imposed a substantial burden on medical resources. This study was triggered by following three factors. First, CVD problem reflects an urgent issue. A high priority been placed long-term therapy and prevention to reduce wastage of resources, particularly developed countries. Second, from perspective preventive medicine, popular data-mining methods have well learned studied, with excellent performance fields. Thus, identification risk factors using these techniques is prime concern. Third, Framingham score core indicator that can be used establish effective prediction model accurately diagnose CVD. this proposes integrated predictive organize five notable classifiers: rough set (RS), decision tree (DT), random forest (RF), multilayer perceptron (MLP), support vector machine (SVM), novel use for attribute selection (i.e., F-attributes first identified study) determine key features identifying Verification experiments were conducted evaluation criteria—accuracy, sensitivity, specificity—based 1190 instances dataset available Taiwan teaching hospital 2019 examples public dataset. Given empirical results, SVM showed best terms accuracy (99.67%), sensitivity (99.93%), specificity (99.71%) all compared other listed classifiers. The RS highest (85.11%), (86.06%), (85.19%) most above results evidence no classifier or suitable practical datasets applications. appropriate address specific data important. Significantly, its calculation attributes DT technique produce entropy-based rules knowledge sets, which not undertaken previous research. conclusively yielded meaningful knowledgeable structures contributed differentiation classifiers two useful research findings helpful management implications subsequent In particular, provide reasonable solutions simplify processes medicine standardizing formats codes problems. significant those past