Enhancing Metabolic Syndrome Prediction with Hybrid Data Balancing and Counterfactuals

2025-04-09

AI Summary

To address class imbalance, limited sample size, and methodological inconsistency in metabolic syndrome (MetS) prediction, this study proposes MetaBoost—a novel end-to-end interpretable machine learning framework. Methodologically, it introduces a weighted iterative synthetic strategy integrating SMOTE, ADASYN, and CTGAN for robust data augmentation, coupled with a MetS-specific clinically interpretable counterfactual reasoning model to quantify biomarker intervention thresholds. The framework employs XGBoost, Random Forest, and TabNet, augmented by random oversampling (ROS), Bayesian probabilistic analysis, and counterfactual generation. Experimental results demonstrate that MetaBoost improves prediction accuracy by 1.14% over baselines. Fasting glucose (50.3%) and triglycerides (46.7%) are identified as the most sensitive intervention features, achieving posterior predictive probabilities of 85.5% and 74.9%, respectively—substantially enhancing clinical decision support.

Technology Category

• Search and Optimization: Metareasoning and Metaheuristics
• Machine Learning: Ensemble Methods
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Metabolic Syndrome (MetS) is a cluster of interrelated risk factors that significantly increases the risk of cardiovascular diseases and type 2 diabetes. Despite its global prevalence, accurate prediction of MetS remains challenging due to issues such as class imbalance, data scarcity, and methodological inconsistencies in existing studies. In this paper, we address these challenges by systematically evaluating and optimizing machine learning (ML) models for MetS prediction, leveraging advanced data balancing techniques and counterfactual analysis. Multiple ML models, including XGBoost, Random Forest, TabNet, etc., were trained and compared under various data balancing techniques such as random oversampling (ROS), SMOTE, ADASYN, and CTGAN. Additionally, we introduce MetaBoost, a novel hybrid framework that integrates SMOTE, ADASYN, and CTGAN, optimizing synthetic data generation through weighted averaging and iterative weight tuning to enhance the model's performance (achieving a 1.14% accuracy improvement over individual balancing techniques). A comprehensive counterfactual analysis is conducted to quantify feature-level changes required to shift individuals from high-risk to low-risk categories. The results indicate that blood glucose (50.3%) and triglycerides (46.7%) were the most frequently modified features, highlighting their clinical significance in MetS risk reduction. Additionally, probabilistic analysis shows elevated blood glucose (85.5% likelihood) and triglycerides (74.9% posterior probability) as the strongest predictors. This study not only advances the methodological rigor of MetS prediction but also provides actionable insights for clinicians and researchers, highlighting the potential of ML in mitigating the public health burden of metabolic syndrome.

Problem

• Addressing class imbalance and data scarcity in MetS prediction
• Optimizing ML models with hybrid data balancing techniques
• Identifying key features for MetS risk reduction via counterfactuals

Innovation

• Hybrid framework MetaBoost combines SMOTE, ADASYN, CTGAN
• Counterfactual analysis identifies key MetS risk features
• Optimized synthetic data generation improves model accuracy