Integrating Nutritional Status in Machine Learning Predictive Models for Cardiovascular Risk: A Pilot Study

This study explored the use of machine learning (ML) models for cardiovascular risk stratification in an elderly Thai population. A cross-sectional analysis was performed in 210 hypertensive adults aged 60 years and older, using age, sex, systolic blood pressure, smoking status, diabetes mellitus, and body mass index as predictors. Model performance was assessed through 5-fold cross-validation, with the random forest model showing the best overall performance (accuracy = 67.14 ± 10.47%, F1-score = 0.57).