Key result
LGBM and Random Forest algorithms achieve ~99% accuracy in predicting cardiovascular risk.
Why the study?
Efficient early prediction of cardiovascular risk using machine learning algorithms is critical for improving timely diagnosis and preventative strategies.
Do machine learning algorithms accurately predict cardiovascular risk based on patient features?
Observational (n=976)
Do machine learning algorithms accurately predict cardiovascular risk based on patient features?
Machine learning models, specifically Light Gradient Boosting Machine and Random Forest, can predict cardiovascular risk with 99% accuracy using standard clinical features.
ML models should not yet guide risk stratification; leaves open need for prospective external validation.
Cardiovascular Disease (CVD) is a critical global health concern requiring efficient early prediction. This study evaluates Machine Learning (ML) algorithms—Logistic Regression (LR), Light Gradient Boosting Machine (LGBM), Random Forest (RF), Decision Tree (DT), and K-Nearest Neighbors (K-NN)—for forecasting the cardiovascular risk. By analyzing key patient features, such as cholesterol, blood pressure, and lifestyle, these models classify individuals into low or high-risk categories. The findings demonstrate that LGBM and RF achieve superior performance, both reaching 99% accuracy, precision, recall, and F1-score, alongside high ROC-Area Under the Curve (AUC) values. This research provides robust, data-driven tools to enhance the timely diagnosis and support preventative medical strategies, ultimately improving the patient outcomes.
No takes yet. Share an insight, caveat, or question.
Sitharamulu et al. (2025) conducted an observational in Cardiovascular Disease (n=976). Light Gradient Boosting Machine (LGBM) and Random Forest (RF) vs. Logistic Regression, Decision Tree, and K-Nearest Neighbors was evaluated on Model accuracy for cardiovascular risk prediction. Light Gradient Boosting Machine (LGBM) and Random Forest (RF) algorithms achieved 99% accuracy, precision, recall, and F1-score in cross-validation for predicting cardiovascular risk.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: