Enhancing Large Language Models with Retrieval-Augmented Generation: A Radiology-Specific Approach
Dane Weinert, Andreas M. Rauschecker
Abstract
Retrieval-augmented generation (RAG) significantly improves the performance of some large language models on a radiology examination, highlighting the need for further optimization and evaluation of RAG in radiology.
Topics & Concepts
Computer scienceEnglish languageInformation retrievalRadiologyMedicinePsychologyMathematics educationArtificial Intelligence in Healthcare and EducationTopic ModelingRadiomics and Machine Learning in Medical Imaging