EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Augmenting large language models with clinical knowledge graph for personalized perioperative fluid therapy question answering

Jie Song iD, Jinhua Feng, Yuxin Zhang, Cheng Bi, Xin Zheng, Zhichuan Xu, Jiale Du, Mengqiao He, Meng Xiao, Xiaoran Li, Qiongfang Cao, Chi Zhang und 3 weitere

PLOS Digital Health · 11.06.2026 · 5 (6) · e0001474

DOI: 10.1371/journal.pdig.0001474 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Digital Health, online erschienen am 11.06.2026. Verfasst von Jie Song und 14 weiteren Autor:innen. Beteiligte Einrichtung: Joint Laboratory of Artificial Intelligence for Critical Care Medicine, Department of Critical Care Medicine and Institutes for Systems Genetics, Fro…. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Sichuan Province Science and Technology Support Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Personalized perioperative fluid therapy is important for reducing postoperative complications and adverse outcomes. Although large language models (LLMs) show promise in healthcare, their application in fluid therapy remains challenged by hallucinations, limited domain-specific knowledge, and insufficient personalization. To address these limitations, Retrieval-Augmented Generation (RAG) is an effective method, while Knowledge Graphs (KGs) provide more accurate and reliable information. In this paper, we constructed a Personalized Fluid Therapy Knowledge Graph (PFTKG) comprising 6,490 entities and 15,687 relationships, and adapted GraphRAG, a graph-based RAG strategy that employs community detection and recursive summarization to support finding-level retrieval of clinically relevant information. We compared GraphRAG with document-based retrieval-augmented generation (DocRAG) and mainstream prompting strategies, including Vanilla, Chain-of-Thought (CoT), and Reflection-of-Thoughts (RoT), across three LLMs: GPT-4o, Claude Opus 4, and Gemini 2.5 Pro. Performance was evaluated using a 300-question knowledge-based question set and a 262-question retrospective case-based question set derived from 206 abdominal surgery patients. Evaluation included accuracy, honesty, error composition, response length, and response time. On the knowledge-based question set, GraphRAG achieved the highest average accuracy: 96.89% for multiple-choice questions and 66.44% for open-ended questions. On the retrospective case-based question set, GraphRAG also showed the strongest overall performance, with an average accuracy of 71.12%, compared with 62.47% for DocRAG, 54.20% for CoT, 52.67% for Vanilla, and 52.54% for RoT. Adding a “Don’t know” option increased explicit acknowledgment of uncertainty, and GraphRAG reduced context-irrelevant errors compared with DocRAG. These results support GraphRAG as a domain-adapted retrieval strategy for personalized perioperative fluid therapy question answering. By integrating a clinical knowledge graph with hierarchical summarization and finding-level retrieval, it improved answer accuracy and promoted more conservative responses under uncertainty in both knowledge-based and retrospective case-based evaluations, supporting its use in future clinically integrated studies.

Autor:innen und Institutionen

  • Joint Laboratory of Artificial Intelligence for Critical Care Medicine, Department of Critical Care Medicine and Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West Chin…
  • West China School of Nursing, West China Hospital, Sichuan University, Chengdu, China.
  • Department of Biliary Surgery, and Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu, ‌‌China.
  • Department of Neurology, West China Hospital, Sichuan University, Chengdu, China.
  • Information Center, Engineering Research Center of Medical Information Technology, Ministry of Education, West China Hospital, Sichuan University, Chengdu, China.
  • Department of Computer Science and Information Technologies, Iberian Society of Telehealth and Telemedicine, University of A Coruña, A Coruña, Spain.

ORCID-Profile: Jie Song, Bairong Shen

Förderung

  • National Natural Science Foundation of China
  • Sichuan Province Science and Technology Support Program

Zitieren

APA

Song, J., Feng, J., Zhang, Y., Bi, C., Zheng, X., Xu, Z., Du, J., He, M., Xiao, M., Li, X., Cao, Q., Zhang, C., Yang, H., Wu, R., & Shen, B. (2026). Augmenting large language models with clinical knowledge graph for personalized perioperative fluid therapy question answering. PLOS Digital Health, 5(6), e0001474. https://doi.org/10.1371/journal.pdig.0001474

Vancouver

Song J, Feng J, Zhang Y, Bi C, Zheng X, Xu Z, et al. Augmenting large language models with clinical knowledge graph for personalized perioperative fluid therapy question answering. PLOS Digital Health. 2026;5(6):e0001474. doi:10.1371/journal.pdig.0001474

BibTeX
@article{Song2026,
  author = {Song, Jie and Feng, Jinhua and Zhang, Yuxin and Bi, Cheng and Zheng, Xin and Xu, Zhichuan and Du, Jiale and He, Mengqiao and Xiao, Meng and Li, Xiaoran and Cao, Qiongfang and Zhang, Chi and Yang, Hao and Wu, Rongrong and Shen, Bairong},
  title = {Augmenting large language models with clinical knowledge graph for personalized perioperative fluid therapy question answering},
  journal = {PLOS Digital Health},
  year = {2026},
  volume = {5},
  number = {6},
  pages = {e0001474},
  doi = {10.1371/journal.pdig.0001474}
}
Kostenlos starten →