Using secure artificial intelligence agents integrated within the electronic medical record for the evaluation of blood culture appropriateness—Northern California, 2025
Infection Control & Hospital Epidemiology · 2026-02 · 47 (2) · 206-209
DOI: 10.1017/ice.2025.10349 ↗Kurz gefasst
Fachartikel in der Zeitschrift Infection Control & Hospital Epidemiology, online erschienen am 11.11.2025. Verfasst von Guillermo Rodriguez-Nava und 7 weiteren Autor:innen. Beteiligte Einrichtung: Division of Infectious Diseases & Geographic Medicine, Stanford University School of Medicinehttps://ror.org/03mtd9a03, Stanford, CA, USA. Hauptthemen laut MEDLINE: Electronic Health Records, Blood Culture, Artificial Intelligence, Decision Support Systems, Clinical. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 10 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
We evaluated large language model (LLM)-based agents integrated with the electronic medical record to assess blood culture appropriateness. While sensitivity was high, specificity remained low. Performance was shaped by prompt phrasing, sycophantic behavior, and semantic triggers, reflecting both the potential and limitations of LLMs in real-world clinical decision support.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Division of Infectious Diseases & Geographic Medicine, Stanford University School of Medicinehttps://ror.org/03mtd9a03, Stanford, CA, USA.
- Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
- Department of Nursing Information & Informatics, Stanford Health Care, Stanford, CA, USA.
ORCID-Profile: Guillermo Rodriguez-Nava, Timothy Keyes, Nerissa Ambers, Eugenia Miranti, Erika Paola Viana-Cardenas, Wajeeha Tariq, Mindy Marie Sampson, Jorge Luis Salinas
Zitieren
APA
Rodriguez-Nava, G., Keyes, T., Ambers, N., Miranti, E., Viana-Cardenas, E. P., Tariq, W., Sampson, M. M., & Salinas, J. L. (2026). Using secure artificial intelligence agents integrated within the electronic medical record for the evaluation of blood culture appropriateness—Northern California, 2025. Infection Control & Hospital Epidemiology, 47(2), 206-209. https://doi.org/10.1017/ice.2025.10349
Vancouver
Rodriguez-Nava G, Keyes T, Ambers N, Miranti E, Viana-Cardenas EP, Tariq W, et al. Using secure artificial intelligence agents integrated within the electronic medical record for the evaluation of blood culture appropriateness—Northern California, 2025. Infection Control & Hospital Epidemiology. 2026;47(2):206-209. doi:10.1017/ice.2025.10349
BibTeX
@article{RodriguezNava2026,
author = {Rodriguez-Nava, Guillermo and Keyes, Timothy and Ambers, Nerissa and Miranti, Eugenia and Viana-Cardenas, Erika Paola and Tariq, Wajeeha and Sampson, Mindy Marie and Salinas, Jorge Luis},
title = {Using secure artificial intelligence agents integrated within the electronic medical record for the evaluation of blood culture appropriateness—Northern California, 2025},
journal = {Infection Control & Hospital Epidemiology},
year = {2026},
volume = {47},
number = {2},
pages = {206--209},
doi = {10.1017/ice.2025.10349}
}
EUVIMED