Cardiovascular Disease Prediction Through Machine Learning Algorithms

Cardio vascular diseases (CVD) remain a global health concern, necessitating advanced predictive tools for early identification and intervention. CVD is a broad term that encompasses a class of diseases related to the heart and blood vessels. These conditions can range from diseases affecting the heart muscle to those impacting the blood vessels, and they are a leading cause of unhealthy and mortality worldwide. This article presents a comprehensive study on predicting cardiovascular diseases (CVD) using the Random Forest (RF) algorithm. RF’s effectiveness lies in its randomness, which reduces overfitting and increases robustness by selecting random subsets of features at each node split and aggregating predictions from multiple decision trees. Through hyperparameter tuning, the study optimizes RF parameters and analyzes the impact of dataset size on predictive accuracy. Results demonstrate RF’s superior performance with an 88.04% accuracy rate, surpassing other classifiers such as XG Boost, SVM, and KNN. Moreover, the study examines the Receiver Operating Characteristics (ROC) curve and Area Under the Curve (AUC) score, affirming RF’s promising performance compared to other algorithms. Further evaluation against existing algorithms, GN-ANN and EHV, across varying dataset sizes reinforces RF’s superiority, indicating its potential for accurate CVD prediction.