Integrating Knowledge Graphs and Bayesian Networks: A Hybrid Approach for Explainable Disease Risk Prediction

2025-06-16

AI Summary

To address the challenges of incomplete electronic health record (EHR) data, uncertain health outcomes, and limited clinical interpretability, this paper proposes a disease risk prediction framework integrating knowledge graphs (KGs) and Bayesian networks (BNs). Methodologically, it introduces the first approach to automatically construct patient-context-aware Bayesian networks from medical ontology-based knowledge graphs, synergistically combining KG embedding, ontology reasoning, multimodal EHR feature learning, and structured probabilistic inference. The framework enables dynamic alignment between general medical knowledge and domain-specific clinical scenarios. Evaluated on atrial fibrillation prediction, it achieves an AUC of 0.89. Moreover, it supports fine-grained causal attribution and counterfactual explanation, substantially enhancing model transparency, clinical trustworthiness, and decision traceability.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Reasoning under Uncertainty: Graphical Models

Application Category

• 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
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Multimodal electronic health record (EHR) data is useful for disease risk prediction based on medical domain knowledge. However, general medical knowledge must be adapted to specific healthcare settings and patient populations to achieve practical clinical use. Additionally, risk prediction systems must handle uncertainty from incomplete data and non-deterministic health outcomes while remaining explainable. These challenges can be alleviated by the integration of knowledge graphs (KGs) and Bayesian networks (BNs). We present a novel approach for constructing BNs from ontology-based KGs and multimodal EHR data for explainable disease risk prediction. Through an application use case of atrial fibrillation and real-world EHR data, we demonstrate that the approach balances generalised medical knowledge with patient-specific context, effectively handles uncertainty, is highly explainable, and achieves good predictive performance.

Problem

• Adapt general medical knowledge to specific healthcare settings
• Handle uncertainty from incomplete data and outcomes
• Integrate KGs and BNs for explainable risk prediction

Innovation

• Integrates knowledge graphs with Bayesian networks
• Uses ontology-based KGs and multimodal EHR data
• Balances general knowledge with patient-specific context