Leveraging Large Language Models in Radiology Research: A Comprehensive User Guide

Abstract

Large Language Models (LLMs) such as ChatGPT have been increasingly integrated into radiology research, revolutionizing the research landscape. The Radiology Research Alliance (RRA) of the Association for Academic Radiology (AAR) has convened a Task Force to develop this guide to help radiology researchers responsibly adopt LLM technologies. LLMs can improve various phases of the research process by helping to automate literature reviews, generate research questions, analyze complex datasets, and draft manuscripts. Despite its potential to improve research efficiency, implementation of LLMs poses challenges, especially for users with limited artificial intelligence (AI) experience. This review focuses on approaches to using LLMs in each phase of the research process and addresses prompt engineering to improve interaction with LLMs as well as ethical concerns to ensure scientific integrity. By combining human expertize with AI-driven efficiency, radiology researchers can foster innovation, advance knowledge, and enhance patient care.