A Systematic Literature Review on Machine Learning Techniques for Heart Disease Prediction

Machine learning (ML) is a fast growing topic of study. A rising number of studies use ML as a powerful theoretical framework. However, there is a paucity of study into predicting the early symptoms of cardiac disease. Improving patient outcomes depends on early diagnosis of cardiac disease. A comprehensive review of several ML algorithms for the prediction of heart disease is discussed in this work. Various methods were reviewed such as Naive Bayes classifiers, regression models (Logistic, Linear, Lasso), Decision trees, Support vector machines, Ensemble methods and Neural networks. The study evaluated the effect of prediction accuracy on Obesity., high blood pressure, diabetes, high cholesterol., alcohol consumption and smoking are important risk factors. These results show that prediction accuracy is greatly improved when ML algorithms are combined with different risk factors. The selection of suitable algorithms to enhance cardiac disease prediction and overall healthcare outcomes are discussed in this review.