Authors - Brajesh Kumar, Harsh Mathur Abstract - The early prediction of cardiovascular disease (CVD) saves millions of lives worldwide. The early symptoms of cardiovascular disease are usual and cannot predict. Atherosclerosis is a significant contributor to cardiovascular disease. Atherosclerosis hardens the coronary artery, reduces blood flow, and increases severe heart attack and stroke. The automated detection of cardiovascular diseases plays an essential role in heart disease. The machine learning-based classification algorithm plays a vital role in accurate classification and detection. This paper proposed ensemble-based classifier methods prediction of atherosclerosis disease. The proposed ensemble classifier built on the principle of boosting methods and use two classifier support vector machine and KNN. the support vector methods work as a base classifier and KNN as a variable classifier. The ensemble classifier works with feature optimization methods. The Glowworm optimization (GSO) algorithm applied for the optimization of features. The optimized features increase the rate of classification accuracy. The proposed algorithm simulated in MATLAB software and tested with three reputed datasets of atherosclerosis. The proposed algorithm compares with the existing three algorithms for atherosclerosis disease detection: support vector machine, KNN and decision tree (DT). The difference of results suggests that the proposed algorithm is efficient instead of the existing algorithms.
Open Zoom