Reasoning-Enhanced Healthcare Predictions with Knowledge Graph Community Retrieval

2024-10-06

arXiv.org

AI Summary

To address the unreliability of large language models (LLMs) in high-stakes clinical prediction—stemming from hallucination and insufficient fine-grained medical knowledge—this paper proposes a knowledge graph (KG)-based community-level retrieval framework synergistically integrated with LLMs for collaborative reasoning. We innovatively construct a multi-source fused medical KG and design a dynamic, multi-dimensional contextual enhancement retrieval mechanism grounded in community-structured KG modeling. Furthermore, we establish an inference-driven, interpretable prediction paradigm that ensures decision traceability. Evaluated on MIMIC-III and MIMIC-IV, our method achieves substantial improvements: mortality prediction accuracy increases by 10.8–15.0%, and readmission prediction accuracy improves by 12.6–12.7%. The framework significantly enhances both predictive reliability and explanatory transparency, offering a novel paradigm for high-assurance clinical decision support systems.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

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

Large language models (LLMs) have demonstrated significant potential in clinical decision support. Yet LLMs still suffer from hallucinations and lack fine-grained contextual medical knowledge, limiting their high-stake healthcare applications such as clinical diagnosis. Traditional retrieval-augmented generation (RAG) methods attempt to address these limitations but frequently retrieve sparse or irrelevant information, undermining prediction accuracy. We introduce KARE, a novel framework that integrates knowledge graph (KG) community-level retrieval with LLM reasoning to enhance healthcare predictions. KARE constructs a comprehensive multi-source KG by integrating biomedical databases, clinical literature, and LLM-generated insights, and organizes it using hierarchical graph community detection and summarization for precise and contextually relevant information retrieval. Our key innovations include: (1) a dense medical knowledge structuring approach enabling accurate retrieval of relevant information; (2) a dynamic knowledge retrieval mechanism that enriches patient contexts with focused, multi-faceted medical insights; and (3) a reasoning-enhanced prediction framework that leverages these enriched contexts to produce both accurate and interpretable clinical predictions. Extensive experiments demonstrate that KARE outperforms leading models by up to 10.8-15.0% on MIMIC-III and 12.6-12.7% on MIMIC-IV for mortality and readmission predictions. In addition to its impressive prediction accuracy, our framework leverages the reasoning capabilities of LLMs, enhancing the trustworthiness of clinical predictions.

Problem

• Addresses LLM hallucinations in clinical decision support
• Improves retrieval of relevant medical knowledge for predictions
• Enhances accuracy and interpretability of healthcare predictions

Innovation

• Integrates KG community retrieval with LLM reasoning
• Constructs multi-source KG with hierarchical organization
• Dynamic retrieval enriches patient contexts for predictions