Evaluating LLMs’ Potential to Identify Rare Patient Identifiers in Patient Health Records
Studies in Health Technology and Informatics · 2025
DOI: 10.3233/shti250486 ↗Kurz gefasst
Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, 2025. Verfasst von Matúš Falis und 3 weiteren Autor:innen. Beteiligte Einrichtung: Usher Institute, University of Edinburgh, Edinburgh. Hauptthemen laut MEDLINE: Electronic Health Records, Natural Language Processing, Patient Identification Systems, Data Anonymization. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
This study explores the utility of Large Language Models (LLMs) to support finding rare patient record details that could make a patient identifiable. Whilst most research has focused on what we call direct patient identifiers, indirect patient identifiers are not widely addressed. Our evaluation of patient records with mentions of indirect risks predicted by our LLM shows the potential to find these risks automatically. However, many risks highlighted were false positives or did not constitute identifiable risk. More work is needed to understand how we can harness the potential of LLMs as part of our de-identification pipelines for patient health records. Better de-identification of health records is important for safely improving data access and advancing research without compromising confidentiality.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Usher Institute, University of Edinburgh, Edinburgh.
Zitieren
APA
Falis, M., Gruber, F., McInerney, S., & Casey, A. (2025). Evaluating LLMs’ Potential to Identify Rare Patient Identifiers in Patient Health Records. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti250486
Vancouver
Falis M, Gruber F, McInerney S, Casey A. Evaluating LLMs’ Potential to Identify Rare Patient Identifiers in Patient Health Records. Studies in Health Technology and Informatics. 2025. doi:10.3233/shti250486
BibTeX
@article{Falis2025,
author = {Falis, Matúš and Gruber, Franz and McInerney, Samuel and Casey, Arlene},
title = {Evaluating LLMs’ Potential to Identify Rare Patient Identifiers in Patient Health Records},
journal = {Studies in Health Technology and Informatics},
year = {2025},
doi = {10.3233/shti250486}
}
EUVIMED