Coronary Heart Diseases Detection using Stacking Based Ensemble with Emperor Penguin Optimization Approach

A buildup of plaque in coronary arteries, the blood veins that supply oxygen-rich blood to heart, results in Coronary Artery Disease, also known as CAD, coronary or atherosclerotic heart disease, which is a dangerous condition. Globally, it has an impact on millions of individuals. Currently doctors conduct a wide variety of tests and evaluate patients’ medical history to diagnose the disease. Hence a method is needed to diagnose the disease automatically and help in the early treatment of the disease. As time progresses the disease becomes worse. At the moment, angiography is performed to assess the degree and location of cardiac vessel blockage. Several academics have been inspired to apply data mining for CAD diagnosis because it is costly and has a lot of adverse side effects. The literature has employed a variety of characteristics and methods. This study proposes a machine learning and bio inspired algorithm based approach, Stacking Based Ensemble with Emperor Penguin Optimization(SEEPO), which uses a two level stacking with level 1 serving as the base level and level 2 as the meta-level. The input for meta-level ML classifiers is chosen from the predictions of base-level classifiers. The 303 cases in the Z-Alizadeh Sani CHD(Coronary Heart Diseases) dataset that we utilize have all been approved by CAG(Coronary Artery Angiography). Experimental findings show that the suggested model identifies CHD with 94.7% accuracy, 94.8% precision, 94.8% recall, and 94.7% f1-score, respectively.