CUI-Curate: a GraphRAG-based framework for automated clinical concept curation for NLP applications
Journal of the American Medical Informatics Association · 2026
DOI: 10.1093/jamia/ocag110 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, 2026. Verfasst von Victoria Blake und 4 weiteren Autor:innen. Beteiligte Einrichtung: Centre for Big Data Research in Health, University of New South Wales, Randwick, NSW 2031, Australia. Hauptthemen laut MEDLINE: Natural Language Processing, Unified Medical Language System, Information Storage and Retrieval. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Australian Government Research Training Program (RTP) Scholarship, Australin Government National Industry PhD Program Scholarship, Eastern Heart Clinic.
Abstract (Originalsprache)
Background
Clinical named entity recognition tools commonly map free text to Unified Medical Language System (UMLS) Concept Unique Identifiers (CUIs). For many downstream tasks, however, the clinically meaningful unit is not a single CUI but a concept set comprising related synonyms, subtypes, and associated concepts. Constructing these sets is labour-intensive, inconsistently performed, and poorly supported by existing tools.
Methods
We present CUI-Curate, a graph-based retrieval-augmented-generation (GraphRAG) framework for automated UMLS concept set curation. A UMLS knowledge graph was constructed and embedded for semantic retrieval. Candidate CUIs were retrieved using graph-based expansion and then filtered and classified using large language models (GPT-5 and Qwen3-32B). The framework was evaluated on five lexically heterogeneous clinical concepts against manually curated concept sets and gold-standard concept sets.
Results
CUI-Curate produced substantially larger and more complete concept sets than the manual benchmarks. A single retrieval configuration across concepts achieved high recall of definitive concepts with manageable candidate sets. GPT-5 outperformed manual curation for all concepts and retained at least 95% of definitive gold-standard CUIs, while Qwen3-32B achieved comparable but slightly lower performance. Many missed concepts were not observed in 10,000 MIMIC-III notes. CUI-Curate infrastructure and end-to-end processing were inexpensive and stable across runs.
Conclusions
CUI-Curate offers a scalable, reproducible, and cost-efficient approach for generating clinician-reviewable UMLS concept sets tailored to clinical natural language processing and phenotyping applications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Centre for Big Data Research in Health, University of New South Wales, Randwick, NSW 2031, Australia.
- Eastern Heart Clinic, Prince of Wales Hospital, Randwick, NSW 2031, Australia.
- Department of Cardiology, Prince of Wales Hospital, South Eastern Sydney Local Health District, Randwick, NSW 2031, Australia.
- NSW Ambulance Aeromedical Operations, Bankstown Helicopter Base, Bankstown, NSW 2200, Australia.
- Department of Anaesthesia, Saint George Hospital, Kogarah, NSW 2217, Australia.
- School of Clinical Medicine, University of New South Wales, Kingsford, NSW 2033, Australia.
ORCID-Profile: Victoria Blake, Mathew Miller, Blanca Gallego
Förderung
- Australian Government Research Training Program (RTP) Scholarship
- Australin Government National Industry PhD Program Scholarship
- Eastern Heart Clinic
- Australian Government Research Training Program
- Australin Government National Industry
Zitieren
APA
Blake, V., Novak, J., Miller, M., Ooi, S. Y., & Gallego, B. (2026). CUI-Curate: a GraphRAG-based framework for automated clinical concept curation for NLP applications. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocag110
Vancouver
Blake V, Novak J, Miller M, Ooi SY, Gallego B. CUI-Curate: a GraphRAG-based framework for automated clinical concept curation for NLP applications. Journal of the American Medical Informatics Association. 2026. doi:10.1093/jamia/ocag110
BibTeX
@article{Blake2026,
author = {Blake, Victoria and Novak, Jamie and Miller, Mathew and Ooi, Sze-Yuan and Gallego, Blanca},
title = {CUI-Curate: a GraphRAG-based framework for automated clinical concept curation for NLP applications},
journal = {Journal of the American Medical Informatics Association},
year = {2026},
doi = {10.1093/jamia/ocag110}
}
EUVIMED