A Comparative Analysis of Machine Learning Classification Algorithms for Cardiovascular Disease Prediction

Cardiovascular disease is a life-threatening condition accounting for over 17 million deaths globally every year. Several studies have revealed the advancement of machine learning algorithms in predictive modeling and risk assessment, leading to improved chronic disease management and preventive measures. The algorithms analyze complex medical datasets to uncover hidden patterns, trends, and correlations that might be difficult for the medical practitioner, thereby improving the efficiency and accuracy of the medical decision-making process. This study utilizes a cardiovascular disease dataset comprising 70,000 instances to compare the performance of ten classification algorithms in predicting cardiovascular diseases. A new machine learning model is proposed, which performs better than all existing models using the same dataset. The results show the XGBoost classifier as the best model for predicting cardiovascular diseases, with an accuracy of 73.3% and an AUC score of 80%.