Predictive Modelling for Heart Disease Risk Assessment Using Logistic Regression in Machine Learning

Worldwide, cardiovascular disease has remained one of the topmost killers among all diseases. This has stirred a high interest in prognostic models with early detection and prevention processes. The project aims at developing a machine learning model which will estimate the presence of heart disease in an individual and its risk factors based on certain health-related aspects including the age, blood, pressure and cholesterol levels-of an individual. Then, this complete dataset would be divided into train group and test group according to logistic regression model which can understand the relationship between the characteristics with the target variables and reduce prediction errors as much as possible. It is shown that the improved model can predict coronary heart disease with limited accuracy, appropriate for decision making support planning tools in health care facilities. In future, it can also help in healthcare with the intention to experiment new methodologies such as machine learning models feature engineering and a new robust algorithm can improve prediction accuracy and relevance to clinical conditions.