Large Language Models Illuminate a Progressive Pathway to Artificial Healthcare Assistant: A Review

🔔 News

• 💥 [2023/11/06] Our review paper is available at here.
• ✨ [2023/11/03] We create this repository to maintain a paper list on Large Language Models (LLMs) in Medicine.

Introduction

In the fast-evolving landscape of artificial intelligence, Large Language Models (LLMs) have emerged as groundbreaking tools with the potential to emulate complex human linguistic abilities. Their profound impact on healthcare, a field at the crossroads of multifaceted data and intricate decision-making, is of immense interest. This repository delves into the integration challenges and showcases the breadth of LLMs' applications within the medical sphere.

Herein, we offer a curated anthology that navigates through the realm of general-purpose and specialized LLMs, elucidating their roles in enhancing medical research, streamlining clinical operations, and supporting diagnostic processes. We cast a spotlight on multimodal LLMs, championing their sophistication in harmonizing varied data streams such as medical imagery and electronic health records (EHRs) to refine diagnostic precision. Advancing into the frontiers of innovation, we explore LLM-empowered autonomous healthcare agents, scrutinizing their capacity for personalized care and intricate clinical reasoning. Additionally, we present a synthesis of evaluative strategies critical for verifying the dependability and security of LLMs within medical settings.

Our extensive analysis sheds light on the transformative promise LLMs hold for healthcare's future. Yet, we underscore the indispensable call for ongoing refinement and ethical vigilance as precursors to their successful clinical integration.

Please note: This repository's scope is centered on the technological evolution of LLMs in medicine. For insights into clinical deployments and applications of LLMs, we invite you to consult our comprehensive review.

We sincerely value all contributions, whether through pull requests, issue reports, emails, or other forms of communication.

Table of Content (ToC)

• Introduction
• Specialized Medical LLMs
• Multimodal LLMs in Medicine
• LLM-Powered Healthcare Agents
• Evaluation
  • Strategies
• Valuable Resources
  • Related Surveys
    • LLM Techniques
    • LLMs in Medicine
  • Repositories
• Project Maintainers & Contributors
• Citing
• Acknowledgement

Evaluating General-Purpose LLMs in Medicine via Prompting

• [2023/11] Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine Harsha Nori et al. arXiv. [paper]
• [2023/10] Exploring the Boundaries of GPT-4 in Radiology Liu et al. EMNLP 2023 main. [paper]
• [2023/08] Evaluating large language models on medical evidence summarization. Liyan Tang et al. npj Digital Medicine. [paper]
• [2023/07] Evaluating Large Language Models for Radiology Natural Language Processing. Zhengliang Liu et al. arXiv. [paper]
• [2023/07] Advanced prompting as a catalyst: Empowering large language models in the management of gastrointestinal cancers Jiajia Yuan et al. The Innovation Medicine. [paper]
• [2023/04] **Are Large Language Models Ready for He