Improving Cardiovascular Disease Risk Prediction with Machine Learning

The worldwide impact of cardiovascular diseases (CVDs) remains a critical issue, contributing to a significant number of fatalities annually. The primary challenge lies in the timely recognition and timely intervention to mitigate risks and improve outcomes. This study addresses this issue by employing machine learning methodologies to estimate the chances of CVDs, utilizing a dataset from Kaggle comprising information on 1000 patients and 14 key attributes including age, cholesterol levels, and prior heart conditions. The training and evaluation of four machine learning models—Random Forest, Logistic Regression, XGBoost, and Decision Tree—were conducted, with an emphasis on metrics including accuracy, recall, precision, and F1-score. Among these models, the Random Forest algorithm demonstrated superior performance, demonstrated an accuracy of 98.50%, a precision reaching 98.00%, and an F1-score hitting 99.00%. The findings highlight the possible use of machine learning techniques in supporting healthcare practitioners with early diagnosis and preventive care. Future enhancements could include incorporating genetic and lifestyle factors, alongside addressing class imbalance issues, to further improve the predictive accuracy of these models.