Bayesian Event-Based Model for Disease Subtype and Stage Inference

2025-12-03

AI Summary

Existing models for chronic disease subtyping and progression staging suffer from limited robustness due to model misspecification. Method: We propose a Bayesian Event Sequence Model (BESM) that jointly infers disease subtypes, individualized event ordering, and patient-stage assignments within a Bayesian framework—balancing robustness with biological interpretability. By incorporating structured priors and rigorous uncertainty quantification, BESM improves tolerance to model misspecification relative to the widely used SuStaIn model. Contribution/Results: On synthetic data, BESM consistently outperforms SuStaIn in both subtype classification accuracy and event sequence recovery fidelity. In a real-world Alzheimer’s disease cohort, BESM-derived subtypes and progression pathways align more closely with established neuropathological consensus—yielding biologically plausible staging trajectories and mechanistic insights. Thus, BESM provides a robust, interpretable tool for precision staging and pathophysiological dissection of neurodegenerative disorders.

Technology Category

• Machine Learning: Bayesian Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Chronic diseases often progress differently across patients. Rather than randomly varying, there are typically a small number of subtypes for how a disease progresses across patients. To capture this structured heterogeneity, the Subtype and Stage Inference Event-Based Model (SuStaIn) estimates the number of subtypes, the order of disease progression for each subtype, and assigns each patient to a subtype from primarily cross-sectional data. It has been widely applied to uncover the subtypes of many diseases and inform our understanding of them. But how robust is its performance? In this paper, we develop a principled Bayesian subtype variant of the event-based model (BEBMS) and compare its performance to SuStaIn in a variety of synthetic data experiments with varied levels of model misspecification. BEBMS substantially outperforms SuStaIn across ordering, staging, and subtype assignment tasks. Further, we apply BEBMS and SuStaIn to a real-world Alzheimer's data set. We find BEBMS has results that are more consistent with the scientific consensus of Alzheimer's disease progression than SuStaIn.

Problem

• Develops a Bayesian model for disease subtype and stage inference
• Compares robustness of new model against existing method SuStaIn
• Evaluates performance on synthetic and real Alzheimer's disease data

Innovation

• Bayesian event-based model for disease progression
• Robust subtype and stage inference from cross-sectional data
• Outperforms existing model in ordering and assignment tasks