EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

From Text to Knowledge: An End-To-End Extraction Pipeline for Clinical Information

Mário Macedo, Joshua Wiedekopf, Tobias Hillmer, Björn Schreiweis, Sylvia Saalfeld, Hannes Ulrich

Studies in Health Technology and Informatics · 2025

DOI: 10.3233/shti251004 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, 2025. Verfasst von Mário Macedo und 5 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Kiel, Germany. Hauptthemen laut MEDLINE: Natural Language Processing, Data Mining, Electronic Health Records. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study explores the use of Large Language Models (LLMs) in extracting and structuring allergic reaction data from non-English clinical free texts. Leveraging open-source models such as Llama 3.1, Qwen 2.5, and Mistral NeMo, the study utilizes 500 anonymized German discharge letters from the University Hospital Schleswig-Holstein to test an end-to-end workflow. The presented approach extracts allergy information, maps it to SNOMED CT codes, and formats it into HL7 FHIR-compliant resources. Results demonstrate high accuracy in substance detection and encoding, although encoding errors occur due to complex string-matching problems. Reaction identification proves challenging, with mixed performance across models. The study highlights the potential for further improvements through model training on specific terminologies and enhanced prompt engineering. Overall, this proof-of-concept shows the promise of LLMs for varied, domain-specific healthcare tasks, enabling automation and scalability in handling complex clinical data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Natural Language ProcessingData MiningElectronic Health RecordsHumansSystematized Nomenclature of MedicineGermany

Stichwörter der Autor:innen

Clinical DocumentationData IntegrationHL7 FHIRLarge Language ModelsSNOMED CT

Autor:innen und Institutionen

  • Institute for Medical Informatics and Statistics, Kiel University and University Hospital Schleswig-Holstein, Kiel, Germany.
  • Medical Data Integration Center, University Hospital Schleswig-Holstein, Germany.
  • Institute of Medical Biometry and Statistics, Section for Clinical Research-IT, University of Lübeck and University Hospital Schleswig-Holstein, Lübeck, Germany.

Zitieren

APA

Macedo, M., Wiedekopf, J., Hillmer, T., Schreiweis, B., Saalfeld, S., & Ulrich, H. (2025). From Text to Knowledge: An End-To-End Extraction Pipeline for Clinical Information. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti251004

Vancouver

Macedo M, Wiedekopf J, Hillmer T, Schreiweis B, Saalfeld S, Ulrich H. From Text to Knowledge: An End-To-End Extraction Pipeline for Clinical Information. Studies in Health Technology and Informatics. 2025. doi:10.3233/shti251004

BibTeX
@article{Macedo2025,
  author = {Macedo, Mário and Wiedekopf, Joshua and Hillmer, Tobias and Schreiweis, Björn and Saalfeld, Sylvia and Ulrich, Hannes},
  title = {From Text to Knowledge: An End-To-End Extraction Pipeline for Clinical Information},
  journal = {Studies in Health Technology and Informatics},
  year = {2025},
  doi = {10.3233/shti251004}
}
Kostenlos starten →