Enhancing cardiovascular disease prediction: A hybrid machine learning approach integrating oversampling and adaptive boosting techniques

This study presents a novel approach to enhance cardiovascular disease prediction using a hybrid machine learning (ML) model. Leveraging on Synthetic Minority oversampling techniques (SMOTE) and adaptive boosting (AdaBoost), we integrate these methods with prominent classifiers, including Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Extra Tree (ET). Focused on heart rate data as stress level indicators, our objective is to jointly predict cardiovascular disease, thereby addressing the global health challenge of early detection and accurate risk assessments. In response to class imbalance issues in cardiology databases, our hybrid model, which combines SMOTE and AdaBoost, demonstrates promising results. The inclusion of diverse classifiers, such as RF, XGBoost, and ET, enables the model to capture both linear and nonlinear relationships within the heart rate data, significantly enhancing the prediction accuracy. This powerful predictive tool empowers healthcare providers to identify individuals at a high risk for heart disease, thus facilitating timely interventions. This article underscores the pivotal role of ML and hybrid methodologies in advancing health research, particularly in cardiovascular disease prediction. By addressing the class imbalance and incorporating robust algorithms, our research contributes to the ongoing efforts to improve predictive modeling in healthcare. The findings presented here hold significance for medical practitioners and researchers striving for the early detection and prevention of cardiovascular diseases.