Retrieval-Augmented Generation for Large Language Models in Radiology: Another Leap Forward in Board Examination Performance
Rajesh Bhayana, Aly Fawzy, Yangqing Deng, Robert Bleakney, Satheesh Krishna
Abstract
Perplexity (Perplexity AI), which uses web-based retrieval-augmented generation (RAG) and prioritizes authoritative sources, with GPT-4 Turbo (OpenAI) outperformed default GPT-4 Turbo on a radiology board–style examination, illustrating the utility of RAG optimization to generate radiology answers.
Topics & Concepts
MedicineMedical physicsRadiologyArtificial Intelligence in Healthcare and EducationRadiomics and Machine Learning in Medical ImagingRadiology practices and education