Bio-Inspired Intelligence for Adaptive Risk Assessment in Heart Disease A Synergistic

Abstract

Coronary Artery disease (CAD) remains a number one cause of global death, highlighting the need for the creation of accurate and effective prediction fashions for early detection and prevention. This work employs the Heart Disease Prediction dataset, incorporating each authentic capabilities and Context-conscious model (CAM) information, to study the impact of sophisticated feature selection methods in conjunction with machine learning algorithms on the accuracy of CAD prediction. a novel hybrid characteristic selection approach, including Harris Hawks Optimization (HHO) and gray Wolf Optimization (GWO), is offered to decorate characteristic subsets and increase model performance. The studies assesses various machine learning techniques, emphasizing a voting Classifier ensemble that consists of Random forest and decision Tree models. The vote casting Classifier attained an accuracy of 79% in the CAM dataset. the usage of HHO-primarily based characteristic selection on the unique dataset finished an impeccable type accuracy of 100%. The hybrid HHO-GWO function choice approach completed an accuracy of 89%, indicating huge enhancements compared to baseline models. The findings spotlight the effectiveness of nature-inspired optimization strategies, specifically HHO and GWO, in enhancing feature selection for CAD prediction models. those findings underscore the promise of those technology in enhancing early analysis gear for CAD and facilitating greater effective preventive healthcare practices.