Leveraging Machine Learning in Public Health: Predictive Modeling for Cardiovascular Disease Risk

Cardiovascular disease (CVD) continues to be the leading cause of death worldwide, accounting for nearly one-third of all fatalities and posing a significant public health challenge. In 2019, CVD was responsible for approximately 9.6 million deaths among men and 8.9 million among women globally, impacting not only the elderly but also younger individuals aged 30 to 70. These statistics highlight the urgent need for effective preventive measures to manage CVD. This project aims to develop a predictive model to assess individual CVD risk using the "Cardiovascular Disease Dataset" from the Mendeley Data repository. We will use RapidMiner to train a model with high accuracy to predict the likelihood of having CVD by analyzing twelve predictor variables. We compare different predictive models using the confusion matrix and ROC curve. We expect our selected model to predict the likelihood of having CVD with a high accuracy. We will also analyze the R square of selected predictors and train those predictors to be able to determine if concentrating on those predictors alone can reduce the chances of CVD. The project will provide insights into false positives as borderline cases. The predictive model will provide healthcare professionals with actionable insights, enabling early detection and personalized care. This project aligns with broader public health goals by promoting early detection and prevention of CVD.