A Metaheuristic-Driven Machine Learning Framework for Heart Disease Diagnosis Using Crow Search Algorithm

Heart disease is a primary global health concern affecting the heart and blood vessels.Heart disease accounts for the highest number of death rates across the world.It affects men and women of all age.About 32% of global deaths are caused by heart disease or cardiovascular disease.Early diagnosis and proper treatment can prevent millions of deaths.Recent improvements in machine learning help in the accurate diagnosis of heart disease by analyzing medical data.By identifying early signs of heart disease, machine learning can help diagnose the disease's cause.The dataset used here is taken from Kaggle which is a combination of data from multiple sources.This work performs the Crow Search algorithm for feature selection and various ensemble and non-ensemble machine learning techniques for classification.This works uses various key metrics to analyse the result.The result shows that HistGradient Boosting, an ensemble model, has the highest accuracy of 97.56%.The result shows the efficiency of machine learning models in disease prediction.