EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Leveraging Large Language Models for the Extraction of General and Domain-Specific Signs and Symptoms

C Mahony Reategui-Rivera, Stefan Escobar-Agreda, Joseph Finkelstein

Studies in Health Technology and Informatics · 2025

DOI: 10.3233/shti250695 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, 2025. Verfasst von C Mahony Reategui-Rivera und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Informatics, School of Medicine, University of Utah, Salt Lake City, Utah. Hauptthemen laut MEDLINE: Natural Language Processing, Electronic Health Records, Data Mining, Symptom Assessment. 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)

Large language models like GPT-4o, a enhanced variant of GPT-4, have the potential to enhance the extraction of signs and symptoms (S&S) from clinical notes, improving the analysis of unstructured data in electronic health records. This study evaluates the performance of GPT-4o in extracting S&S from clinical notes, focusing on both overall and urinary-specific S&S. Clinical notes from the MTSamples corpus were analyzed using GPT-4o, with precision, recall, and F1-score assessed against human-labeled data. Consistency was measured using agreement rate and Cohen’s Kappa. GPT-4o showed moderate performance in overall S&S extraction (precision: 53.8%, recall: 92.6%, F1-score: 68.1%). For urinary-specific S&S, performance improved significantly, with a perfect precision of 100%, recall of 92.2%, and F1-score of 95.9%. Consistency was high, though variability was noted in agreement metrics. The findings confirm the efficacy of GPT-4o for extracting urinary symptoms and underscore the need for domain-specific approaches to optimize clinical NLP tasks.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Natural Language ProcessingElectronic Health RecordsData MiningSymptom AssessmentHumansLarge Language Models

Stichwörter der Autor:innen

Artificial IntelligenceLarge Language ModelsNamed Entity ExtractionNatural Language ProcessingSignsSymptoms

Autor:innen und Institutionen

  • Department of Biomedical Informatics, School of Medicine, University of Utah, Salt Lake City, Utah.
  • Unidad de Telesalud, Facultad de Medicina, Universidad Nacional Mayor de San Marcos, Lima, Peru.

Zitieren

APA

Reategui-Rivera, C. M., Escobar-Agreda, S., & Finkelstein, J. (2025). Leveraging Large Language Models for the Extraction of General and Domain-Specific Signs and Symptoms. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti250695

Vancouver

Reategui-Rivera CM, Escobar-Agreda S, Finkelstein J. Leveraging Large Language Models for the Extraction of General and Domain-Specific Signs and Symptoms. Studies in Health Technology and Informatics. 2025. doi:10.3233/shti250695

BibTeX
@article{ReateguiRivera2025,
  author = {Reategui-Rivera, C Mahony and Escobar-Agreda, Stefan and Finkelstein, Joseph},
  title = {Leveraging Large Language Models for the Extraction of General and Domain-Specific Signs and Symptoms},
  journal = {Studies in Health Technology and Informatics},
  year = {2025},
  doi = {10.3233/shti250695}
}
Kostenlos starten →