Evaluation of Reliability, Repeatability, Robustness, and Confidence of GPT-3.5 and GPT-4 on a Radiology Board–style Examination
Radiology · 01.05.2024 · 311 (2)
DOI: 10.1148/radiol.232715 ↗Kurz gefasst
Fachartikel in der Zeitschrift Radiology, erschienen am 01.05.2024. Verfasst von Satheesh Krishna und 3 weiteren Autor:innen. Beteiligte Einrichtung: From the University Medical Imaging Toronto, Joint Department of Medical Imaging, University Health Network, Mount Sinai Hospital and Women's College…. Hauptthemen laut MEDLINE: Clinical Competence, Educational Measurement, Radiology, Artificial Intelligence. Die Arbeit wurde bisher 84-mal zitiert (Crossref) und nennt 18 Literaturangaben.
Abstract (Originalsprache)
Default GPT-3.5 and GPT-4 were reliably accurate over three attempts at answering radiology board–style multiple-choice questions but had poor repeatability and robustness and were frequently overconfident, limiting usability without domain-specific optimization.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- From the University Medical Imaging Toronto, Joint Department of Medical Imaging, University Health Network, Mount Sinai Hospital and Women's College Hospital, University of Toronto, Toronto General Hospital, 200 Elizab…
ORCID-Profile: Satheesh Krishna, Nishaant Bhambra, Robert Bleakney, Rajesh Bhayana
Zitieren
APA
Krishna, S., Bhambra, N., Bleakney, R., & Bhayana, R. (2024). Evaluation of Reliability, Repeatability, Robustness, and Confidence of GPT-3.5 and GPT-4 on a Radiology Board–style Examination. Radiology, 311(2). https://doi.org/10.1148/radiol.232715
Vancouver
Krishna S, Bhambra N, Bleakney R, Bhayana R. Evaluation of Reliability, Repeatability, Robustness, and Confidence of GPT-3.5 and GPT-4 on a Radiology Board–style Examination. Radiology. 2024;311(2). doi:10.1148/radiol.232715
BibTeX
@article{Krishna2024,
author = {Krishna, Satheesh and Bhambra, Nishaant and Bleakney, Robert and Bhayana, Rajesh},
title = {Evaluation of Reliability, Repeatability, Robustness, and Confidence of GPT-3.5 and GPT-4 on a Radiology Board–style Examination},
journal = {Radiology},
year = {2024},
volume = {311},
number = {2},
doi = {10.1148/radiol.232715}
}
EUVIMED