Explainable Bayesian Framework for Medical Diagnosis: Asymmetric Decisions, Asymptotic Calibration, and Scalable Inference
DOI:
https://doi.org/10.18687/LACCEI2026.1.1.2473Palabras clave:
Bayesian inference, uncertainty quantification, clinical decision theory, asymptotic calibration, variational inferenceResumen
Automated medical diagnosis requires predictive models that not only achieve high accuracy but also provide principled and interpretable uncertainty quantification aligned with clinical decision-making. In this work, we develop a comprehensive mathematical framework for binary medical classification grounded in Bayesian inference, integrating clinical decision theory under asymmetric loss, strong asymptotic guarantees, and computational analysis. We show that the Bayesian predictor minimizes expected clinical risk under realistic loss functions, is asymptotically calibrated, and satisfies the Bernstein–von Mises theorem under regularity conditions. We introduce the Clinical Uncertainty Index (CI), a theoretically grounded metric that decomposes predictive uncertainty into epistemic and aleatory components. The CI is proven to converge to zero with increasing sample size under correct model specification and to act as a diagnostic indicator of model misspecification. Extending the analysis to approximate inference, we formally demonstrate that mean-field variational approximations systematically underestimate posterior uncertainty, potentially inducing diagnostic overconfidence. These findings highlight the critical role of faithful uncertainty estimation in high-stakes clinical applications. Overall, this study provides rigorous mathematical foundations for the development of explainable and clinically reliable medical AI systems with explicit probabilistic guarantees.Descargas
Publicado
2026-07-27
Número
Sección
Articles
Derechos de autor
Derechos de autor 2026 LACCEI
Licencia
Esta obra está bajo una Licencia Creative Commons Atribución-NoComercial-CompartirIgual 4.0 Internacional.
LACCEI conserva el copyright de todos los artículos publicados bajo los términos de su acuerdo de transferencia de copyright. Como titular del copyright, LACCEI distribuye los artículos al público bajo la Licencia Internacional Creative Commons Atribución-NoComercial-CompartirIgual 4.0 (CC BY-NC-SA 4.0).
Cómo citar
Rojas Peñafiel, J. A., & Valladares Patiño, A. G. (2026). Explainable Bayesian Framework for Medical Diagnosis: Asymmetric Decisions, Asymptotic Calibration, and Scalable Inference. LACCEI, 1(14). https://doi.org/10.18687/LACCEI2026.1.1.2473