Performance of GPT-4 with Vision on Text- and Image-based ACR Diagnostic Radiology In-Training Examination Questions
Nolan Hayden, Spencer Gilbert, Laila Poisson, Brent Griffith, Chad Klochko
Abstract
values of .27 to >.99. Conclusion While GPT-4V demonstrated a level of competence in text-based questions, it showed deficits interpreting radiologic images. © RSNA, 2024 See also the editorial by Deng in this issue.
Topics & Concepts
MedicineMedical physicsRadiologyArtificial intelligenceTraining (meteorology)Computer visionComputer scienceMeteorologyPhysicsRadiology practices and educationArtificial Intelligence in Healthcare and EducationRadiomics and Machine Learning in Medical Imaging