Bayesian--AI Fusion for Epidemiological Decision Making: Calibrated Risk, Honest Uncertainty, and Hyperparameter Intelligence

2025-11-14

AI Summary

Machine learning models in epidemiology often lack well-calibrated uncertainty quantification, while Bayesian methods—though theoretically advantageous—are difficult to integrate into modern AI workflows. Method: We propose a two-tiered Bayesian–AI fusion framework: an upper tier employs Bayesian logistic regression to yield individualized risk estimates with calibrated credible intervals; a lower tier leverages Gaussian process-based Bayesian optimization to jointly tune hyperparameters of a penalized Cox survival model, treating hyperparameter selection as Bayesian inference over a black-box objective. Contribution/Results: The framework unifies calibrated risk prediction and principled hyperparameter optimization. Evaluated on diabetes and breast cancer datasets, it significantly improves AUC, Brier score, and log loss; achieves reliable coverage probabilities; and attains calibration consistency approaching the oracle. This enhances predictive reliability and clinical decision-support utility.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Probabilistic Programming
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Modern epidemiological analytics increasingly use machine learning models that offer strong prediction but often lack calibrated uncertainty. Bayesian methods provide principled uncertainty quantification, yet are viewed as difficult to integrate with contemporary AI workflows. This paper proposes a unified Bayesian and AI framework that combines Bayesian prediction with Bayesian hyperparameter optimization. We use Bayesian logistic regression to obtain calibrated individual-level disease risk and credible intervals on the Pima Indians Diabetes dataset. In parallel, we use Gaussian-process Bayesian optimization to tune penalized Cox survival models on the GBSG2 breast cancer cohort. This yields a two-layer system: a Bayesian predictive layer that represents risk as a posterior distribution, and a Bayesian optimization layer that treats model selection as inference over a black-box objective. Simulation studies in low- and high-dimensional regimes show that the Bayesian layer provides reliable coverage and improved calibration, while Bayesian shrinkage improves AUC, Brier score, and log-loss. Bayesian optimization consistently pushes survival models toward near-oracle concordance. Overall, Bayesian reasoning enhances both what we infer and how we search, enabling calibrated risk and principled hyperparameter intelligence for epidemiological decision making.

Problem

• Calibrating uncertainty in machine learning models for epidemiological predictions
• Integrating Bayesian methods with contemporary AI workflows effectively
• Optimizing hyperparameter selection through Bayesian inference techniques

Innovation

• Combines Bayesian prediction with hyperparameter optimization
• Uses Bayesian logistic regression for calibrated risk
• Applies Gaussian-process Bayesian optimization for model tuning