FEDERATED LARGE LANGUAGE MODELS IN HEALTHCARE

A SYSTEMATIC REVIEW, OPPORTUNITIES AND CHALLENGES

Abstract

The convergence of Federated Learning (FL) and Large Language Models (LLMs) represents a transformative opportunity in healthcare. FL allows decentralized model training across multiple institutions without sharing sensitive data, which is crucial in the privacy-sensitive domain of healthcare. Meanwhile, with their exceptional natural language processing (NLP) capabilities, Large Language Models (LLMs) have demonstrated outstanding potential in healthcare applications such as clinical documentation, decision support, and patient record analysis. Despite growing interest in FL and LLM within the healthcare sector, there remains a notable gap in the literature regarding a holistic examination of these technologies opportunities, challenges, and practical applications in the healthcare context. This systematic review synthesizes cutting-edge research and identifies gaps in recent advances in combining FL and LLMs within healthcare, outlining key opportunities and challenges. This review serves as both a synthesis of current knowledge and a roadmap for future research to enable secure, collaborative, and equitable AI-driven healthcare.