OpenTCM: A GraphRAG-Empowered LLM-based System for Traditional Chinese Medicine Knowledge Retrieval and Diagnosis

TL;DR

OpenTCM is an innovative system that leverages a large TCM knowledge graph and GraphRAG to improve knowledge retrieval and diagnosis in Traditional Chinese Medicine, addressing challenges of complex texts and semantic relationships.

Contribution

This work introduces a comprehensive TCM knowledge graph and integrates GraphRAG with LLMs for improved retrieval and diagnosis without model fine-tuning.

Findings

Achieved high expert scores in ingredient retrieval and diagnosis tasks.

Outperformed state-of-the-art solutions in TCM knowledge retrieval.

Constructed a large, multi-relational TCM knowledge graph with high semantic fidelity.

Abstract

Traditional Chinese Medicine (TCM) represents a rich repository of ancient medical knowledge that continues to play an important role in modern healthcare. Due to the complexity and breadth of the TCM literature, the integration of AI technologies is critical for its modernization and broader accessibility. However, this integration poses considerable challenges, including the interpretation of obscure classical Chinese texts and the modeling of intricate semantic relationships among TCM concepts. In this paper, we develop OpenTCM, an LLM-based system that combines a domain-specific TCM knowledge graph and Graph-based Retrieval-Augmented Generation (GraphRAG). First, we extract more than 3.73 million classical Chinese characters from 68 gynecological books in the Chinese Medical Classics Database, with the help of TCM and gynecology experts. Second, we construct a comprehensive multi-relational knowledge graph comprising more than 48,000 entities and 152,000 interrelationships, using customized prompts and Chinese-oriented LLMs such as DeepSeek and Kimi to ensure high-fidelity semantic understanding. Last, we empower OpenTCM with GraphRAG, enabling high-fidelity ingredient knowledge retrieval and diagnostic question-answering without model fine-tuning. Experimental evaluations demonstrate that OpenTCM achieves mean expert scores (MES) of 4.378 in ingredient information retrieval and 4.045 in diagnostic question-answering tasks, outperforming state-of-the-art solutions in real-world TCM use cases.