Analysis of Cardiovascular Diseases Prediction Using Machine Learning Classification Algorithms

Worldwide healthcare systems have faced enormous hurdles because of the COVID-19 pandemic, especially when it comes to treating individuals who already have pre-existing disorders such as cardiovascular diseases (CVDs). Prioritizing medical therapies and resources for COVID-19 patients who are at increased risk of mortality from underlying CVDs requires early identification. In this work, we investigate how well three machine learning algorithms-, Random Forest, XGBoost, and Logistic Regression-predict death in COVID-19 patients who already have cardiovascular disease. We performed grid search and cross-validation using a dataset of clinical and demographic features of COVID-19 patients with and without CVDs to reduce overfitting and maximize model performance. Our findings show that among patients with CVDs, Logistic Regression had the best accuracy in predicting COVID-19 fatality, followed by Random Forest and Decision Tree coming in a close second. These results highlight how machine learning algorithms can help clinical professionals detect high-risk COVID-19 patients who have underlying cardiovascular diseases (CVDs), enable prompt interventions, and enhance patient outcomes.