EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TCMRD - KG: innovative design and development of rheumatology knowledge graph in ancient Chinese literature assisted by large language models

Haotian Li, Congmin Xia, Youjuan Hou, Sile Hu, Yanjun Liu, Quan Jiang

Frontiers in Pharmacology · 19.02.2025 · 16

DOI: 10.3389/fphar.2025.1535596 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Pharmacology, online erschienen am 19.02.2025. Verfasst von Haotian Li und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, China. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Rheumatic immune diseases are a type of immune-inflammatory disease that affects muscles, bones, joints, and surrounding soft tissues. They have a long course and a high disability rate, seriously affecting the quality of life of patients. Traditional Chinese medicine plays an important role in the diagnosis and treatment of rheumatic immune diseases. The unique theoretical system and rich treatment methods of traditional Chinese medicine are preserved in ancient Chinese medical books.

Methods

This study takes the content related to rheumatism in ancient traditional Chinese medicine books as the research object, integrates ontology theory and technology into the knowledge graph, and realizes the reconstruction of traditional Chinese medicine information knowledge. It provides a basic data structure for data mining and knowledge discovery.

Results

This study is the first rheumatism-specific knowledge graph constructed based on ancient traditional Chinese medicine books. It has explored the construction method of a knowledge graph from ancient books by combining automatic labeling of mainstream large language models with manual review. Considering the knowledge characteristics of ancient traditional Chinese medicine books, where existing word segmentation technology struggles to accurately reproduce the original meaning, a new type of entity extraction method is proposed.

Discussion

This provides an important foundation for improving the clinical diagnosis and treatment level of traditional Chinese medicine in treating rheumatism, further exploring the knowledge representation and application of traditional Chinese medicine in rheumatism treatment, and it has potential for future expansion and improvement.

Schlagwörter

Stichwörter der Autor:innen

LLMSartifcial intelligencedata managamentknowledge graph (KG)rheumatictrad. Chinese medicine (TCM)

Autor:innen und Institutionen

  • School of Traditional Chinese Medicine, Beijing University of Chinese Medicine, Beijing, China.
  • Guang'anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China.
  • Institute of Information on Traditional Chinese Medicine, China Academy of Chinese Medical Sciences, Beijing, China.
  • Institute of Basic Theory for Chinese Medicine, China Academy of Chinese Medical Sciences, Beijing, China.

Zitieren

APA

Li, H., Xia, C., Hou, Y., Hu, S., Liu, Y., & Jiang, Q. (2025). TCMRD - KG: innovative design and development of rheumatology knowledge graph in ancient Chinese literature assisted by large language models. Frontiers in Pharmacology, 16. https://doi.org/10.3389/fphar.2025.1535596

Vancouver

Li H, Xia C, Hou Y, Hu S, Liu Y, Jiang Q. TCMRD - KG: innovative design and development of rheumatology knowledge graph in ancient Chinese literature assisted by large language models. Frontiers in Pharmacology. 2025;16. doi:10.3389/fphar.2025.1535596

BibTeX
@article{Li2025,
  author = {Li, Haotian and Xia, Congmin and Hou, Youjuan and Hu, Sile and Liu, Yanjun and Jiang, Quan},
  title = {TCMRD - KG: innovative design and development of rheumatology knowledge graph in ancient Chinese literature assisted by large language models},
  journal = {Frontiers in Pharmacology},
  year = {2025},
  volume = {16},
  doi = {10.3389/fphar.2025.1535596}
}
Kostenlos starten →