Predictive Modeling of Weight Loss and Metabolic Health Outcomes: A Retrospective Predictive Modeling Study

Machine learning models, especially ensemble methods like Stacking and XGBoost, effectively predict body weight loss and MetS improvement in obese patients following a Mediterranean diet. Key predictors include age, insulin resistance markers, and inflammatory biomarkers. Integrating these predictive tools can significantly enhance personalized dietary interventions, optimizing treatment outcomes in clinical practice.