ADVANCING CARDIAC DISEASE PREDICTION VIA HYBRID ML STRATEGIES

Abstract

Researchers have paid close consideration to the field of healthcare research. A large number of researchers have found several reasons for human early death. The associated research in the past has established that illnesses are triggered by a variety of factors, one of which is heart disease. Several investigators advocated unconventional approaches for preserving human life and assisting healthcare professionals in recognizing, preventing, and managing cardiac disease. A few handy approaches aid the professional's judgement; however, each effective plan has a unique set of constraints. Consequently, we are going to develop cardiac disease prediction framework with the deployment of hybrid (stacking) model along with the common machine learning methods like Decision Tree, Random Forest, and XGBoost. The selection of the models plays a crucial part in predicting the heart disease-associated risks, which could then cause stroke in human beings. Finally, the performance was validated using the simulation and the outputs were presented to exhibit the prediction scope of our framework.