Empowering Digital Transformation of Traditional Chinese Medicine with Knowledge Graph-Enhanced Large Language Models: Opportunities and Challenges

Within the new paradigm of digital transformation, the convergence of Knowledge Graph (KG) and Large Language Models (LLM) presents new opportunities to enhance Traditional Chinese Medicine (TCM). Based on literature keyword co-occurrence networks analysis via VOSviewer and multi-case studies, this research discusses the potential and difficulties of KG-LLM implementation in TCM. Analytical outcomes indicate the exceptional efficiency of KG-LLM systems in intelligent diagnosis and treatment, drug discovery, and inheritance learning scenarios. Technical deployment is faced with multi-sided challenges: multi-modal alignment challenges owing to multi-source heterogeneous vocabularies of TCM, clinical credibility loss on account of the black-box nature of LLMs, and technical iteration constraints due to delayed dynamic updating of knowledge and obstacles of interdisciplinary collaboration. Future research will need to concentrate on terminology standardization, development of dynamic knowledge platforms, model explainability, and optimal multi-center validation systems. These developments will allow systematic innovation of TCM's evolution from empirical roots to digital paradigms and build technological infrastructure for international development.