EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

How to benchmark medical AI agents

Silas Ruhrberg Estévez iD, Dyke Ferber iD, Mihaela van der Schaar, Jakob Nikolas Kather iD

PLOS Medicine · 09.07.2026 · 23 (7) · e1005170

DOI: 10.1371/journal.pmed.1005170 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Medicine, online erschienen am 09.07.2026. Verfasst von Silas Ruhrberg Estévez und 3 weiteren Autor:innen. Beteiligte Einrichtung: Else Kroener Fresenius Center for Digital Health, Faculty of Medicine and University Hospital Carl Gustav Carus, TUD Dresden University of Technology…. Hauptthemen laut MEDLINE: Artificial Intelligence, Benchmarking. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 12 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Medical artificial intelligence research is shifting from single-task models toward multimodal large language model-based agents for complex clinical workflows, requiring benchmarks that assess clinical reasoning, process safety, and resource stewardship rather than final outputs alone.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceBenchmarkingHumansLarge Language Models

Autor:innen und Institutionen

  • Else Kroener Fresenius Center for Digital Health, Faculty of Medicine and University Hospital Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany.
  • Cambridge Centre for AI in Medicine, University of Cambridge, Cambridge, United Kingdom.
  • Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, United Kingdom.
  • Department of Medicine I, Faculty of Medicine and University Hospital Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany.
  • Medical Oncology, National Center for Tumor Diseases (NCT), University Hospital Heidelberg, Heidelberg, Germany.
  • Pathology & Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, United Kingdom.

ORCID-Profile: Silas Ruhrberg Estévez, Dyke Ferber, Jakob Nikolas Kather

Zitieren

APA

Ruhrberg Estévez, S., Ferber, D., van der Schaar, M., & Kather, J. N. (2026). How to benchmark medical AI agents. PLOS Medicine, 23(7), e1005170. https://doi.org/10.1371/journal.pmed.1005170

Vancouver

Ruhrberg Estévez S, Ferber D, van der Schaar M, Kather JN. How to benchmark medical AI agents. PLOS Medicine. 2026;23(7):e1005170. doi:10.1371/journal.pmed.1005170

BibTeX
@article{RuhrbergEstvez2026,
  author = {Ruhrberg Estévez, Silas and Ferber, Dyke and van der Schaar, Mihaela and Kather, Jakob Nikolas},
  title = {How to benchmark medical AI agents},
  journal = {PLOS Medicine},
  year = {2026},
  volume = {23},
  number = {7},
  pages = {e1005170},
  doi = {10.1371/journal.pmed.1005170}
}
Kostenlos starten →