Evaluating the Performance and Clinical Utility of AI-driven Diagnostic Tools in Radiology
Radiology · 01.11.2025 · 317 (2)
DOI: 10.1148/radiol.243935 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift Radiology, erschienen am 01.11.2025. Verfasst von Sarah F Mercaldo und 2 weiteren Autor:innen. Beteiligte Einrichtung: Digital CRO, Mass General Brigham AI, 399 Revolution Dr, Somerville, MA 02145. Hauptthemen laut MEDLINE: Artificial Intelligence, Radiology, Diagnostic Imaging. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 95 Literaturangaben.
Abstract (Originalsprache)
This article provides a practical guide for rigorously evaluating artificial intelligence–driven diagnostic tools in radiology, emphasizing essential statistical methods, testing processes, common pitfalls, and real-world clinical applicability to ensure accurate, reliable, and meaningful integration into clinical practice.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Digital CRO, Mass General Brigham AI, 399 Revolution Dr, Somerville, MA 02145.
- Department of Radiology, Massachusetts General Hospital, Boston, Mass.
- Harvard Medical School, Boston, Mass.
- Department of Neurology, Massachusetts General Hospital, Boston, Mass.
- School of Data Science, University of Virginia, Charlottesville, Va.
ORCID-Profile: Sarah F Mercaldo, James M Hillis
Zitieren
APA
Mercaldo, S. F., Hillis, J. M., & Blume, J. D. (2025). Evaluating the Performance and Clinical Utility of AI-driven Diagnostic Tools in Radiology. Radiology, 317(2). https://doi.org/10.1148/radiol.243935
Vancouver
Mercaldo SF, Hillis JM, Blume JD. Evaluating the Performance and Clinical Utility of AI-driven Diagnostic Tools in Radiology. Radiology. 2025;317(2). doi:10.1148/radiol.243935
BibTeX
@article{Mercaldo2025,
author = {Mercaldo, Sarah F and Hillis, James M and Blume, Jeffrey D},
title = {Evaluating the Performance and Clinical Utility of AI-driven Diagnostic Tools in Radiology},
journal = {Radiology},
year = {2025},
volume = {317},
number = {2},
doi = {10.1148/radiol.243935}
}
EUVIMED