Clinical Insights: A Comprehensive Review of Language Models in Medicine

TL;DR

This paper provides a comprehensive synthesis of language models in medicine, tracing the field from encoder-based systems to modern large and multimodal models, with a strong emphasis on locally deployable solutions for privacy and autonomy. It presents a structured taxonomy of clinical NLP tasks, evaluates current approaches across text generation, classification, QA/IE, summarization, and conversation, and discusses the challenges of evaluation, ethics, datasets, and human-AI collaboration. Key contributions include a tiered ethical framework, guidance on task-oriented evaluation, and a call for open, domain-specific datasets and Medical Model Cards to facilitate safe, real-world adoption. The work highlights the practical impact of LM-based tools for documentation, decision support, patient communication, and mental health, while underscoring the need for empirical studies and responsible deployment practices in healthcare settings.

Abstract

This paper explores the advancements and applications of language models in healthcare, focusing on their clinical use cases. It examines the evolution from early encoder-based systems requiring extensive fine-tuning to state-of-the-art large language and multimodal models capable of integrating text and visual data through in-context learning. The analysis emphasizes locally deployable models, which enhance data privacy and operational autonomy, and their applications in tasks such as text generation, classification, information extraction, and conversational systems. The paper also highlights a structured organization of tasks and a tiered ethical approach, providing a valuable resource for researchers and practitioners, while discussing key challenges related to ethics, evaluation, and implementation.