A Retrieval-Augmented Knowledge Mining Method with Deep Thinking LLMs for Biomedical Research and Clinical Support

2025-03-29

AI Summary

To address challenges in biomedical knowledge graph (KG) construction—including terminological complexity, data heterogeneity, dynamic knowledge evolution, and cross-document reasoning—this paper proposes IP-RAR, an integrated retrieval–reasoning framework. IP-RAR introduces a novel two-stage paradigm combining retrieval with self-reflective, deep reasoning to overcome bottlenecks in multi-hop inference and implicit knowledge recall. Leveraging this framework, we construct BioStrataKG—a dynamically evolving biomedical KG—and BioCDQA, a cross-document question-answering benchmark. We further implement LLM-powered automated KG construction, RAG-enhanced generation, and multi-hop semantic matching. Experiments demonstrate a 20% improvement in document retrieval F1-score and a 25% gain in answer accuracy. The framework effectively supports clinical decision-making for personalized medication and facilitates research trend analysis and gap identification.

Technology Category

• Knowledge Representation and Reasoning: Knowledge Acquisition
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Knowledge graphs and large language models (LLMs) are key tools for biomedical knowledge integration and reasoning, facilitating structured organization of scientific articles and discovery of complex semantic relationships. However, current methods face challenges: knowledge graph construction is limited by complex terminology, data heterogeneity, and rapid knowledge evolution, while LLMs show limitations in retrieval and reasoning, making it difficult to uncover cross-document associations and reasoning pathways. To address these issues, we propose a pipeline that uses LLMs to construct a biomedical knowledge graph (BioStrataKG) from large-scale articles and builds a cross-document question-answering dataset (BioCDQA) to evaluate latent knowledge retrieval and multi-hop reasoning. We then introduce Integrated and Progressive Retrieval-Augmented Reasoning (IP-RAR) to enhance retrieval accuracy and knowledge reasoning. IP-RAR maximizes information recall through Integrated Reasoning-based Retrieval and refines knowledge via Progressive Reasoning-based Generation, using self-reflection to achieve deep thinking and precise contextual understanding. Experiments show that IP-RAR improves document retrieval F1 score by 20% and answer generation accuracy by 25% over existing methods. This framework helps doctors efficiently integrate treatment evidence for personalized medication plans and enables researchers to analyze advancements and research gaps, accelerating scientific discovery and decision-making.

Problem

• Constructing biomedical knowledge graphs with complex terminology and data heterogeneity
• Improving retrieval and reasoning capabilities of LLMs in biomedical contexts
• Enhancing cross-document associations and multi-hop reasoning for clinical support

Innovation

• LLMs construct biomedical knowledge graph BioStrataKG
• Integrated Progressive Retrieval-Augmented Reasoning IP-RAR
• Self-reflection enhances retrieval and reasoning accuracy