American Journal for Young Scientists ยท ISSN 3068-9015
A Comparative Analysis of the Machine Learning Models for Cardiovascular Risk Prediction
Authors: Akshita Vivek
Published: October 15, 2025
Category: Biology
Abstract
Cardiovascular diseases (CVDs) are among the leading causes of death worldwide and represent one of the significant public health concerns, highlighting the importance of early diagnosis for effective management and treatment. Traditional methods for predicting CVD risks have limitations and fail to capture the complex interactions among various risk factors that influence disease progression. In this study, we employed machine learning (ML) and artificial intelligence (AI) models, such as extreme gradient boosting (XGBoost), multi-layer perceptron (MLP), support vector machine (SVM), and linear discriminant analysis (LDA), on two publicly available CVD datasets to evaluate the accuracy of risk prediction for cardiovascular diseases. The results showed that the XGBoost and MLP models outperformed the SVM and LDA models in predicting CVD risks. Remarkably, with the heart disease dataset, we achieved 100% accuracy in predicting the CVD risk using XGBoost and MLP models. In this dataset, with both XGBoost and MLP models, the accuracy, precision, recall values, and AUC were 1.0. The cardiovascular disease dataset showed accuracy, precision, recall values of 0.99 and an AUC of 1.0 with the XGBoost model, while the MLP model had these values: 0.97, 0.95, 0.99, and 0.99, respectively. When comparing our results with other published research on machine learning approaches for predicting CVD risks across diverse CVD datasets, our results outperformed all other models, further emphasizing the significance of this study. These findings suggest that ML-driven CVD risk assessment could be integrated into clinical decision-support systems to help physicians in early intervention.