Heart Disease Prediction using Machine Learning

Heart attack and other heart diseases have acted as the major global cause of mortality. In accordance with WHO, about 32% death was due to cardiovascular disease, among which 85% was only due to heart disease There are various factors that lead to heart disease including high LDL, cholesterol, high blood pressure, Diabetes Smoking, Unhealthy diet, Obesity and Physical inactivity. If heart disease is not predicted earlier, it reduces patient’s chances of survival. Therefore, it is essential to design a secure and precise system to predict heart disease early for proper treatment of patients. This research paper presents comprehensive analysis to identify heart disease using different predictive analytic methods. The research utilized a dataset of 1025 instances with 13 input parameters compromising age, sex, chest pain, blood pressure, blood sugar, electrocardiographic results, heart rate etc. The research tests and implements various predictive analytic methods such as Decision tree, Random Forest, Logistic regression and KN neighbor to predict the risk of heart disease on 60:40, 70:30 and 80:20 ratio data split. This research paper also highlights the superiority of Random Forest model for predicting heart disease with highest accuracy.