Previously submitted to: Journal of Medical Internet Research (no longer under consideration since Sep 09, 2025)
Date Submitted: Jan 9, 2025
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Revolutionizing Dermatology: Opportunities and Challenges of Large Language Models
ABSTRACT
Background:
The diagnosis and treatment of dermatological conditions heavily rely on physicians' expertise, with inconsistencies in care quality at the primary level. Recent advancements in large language models (LLMs) have shown promise in medical knowledge extraction and intelligent consultation. However, challenges such as training data biases and limited model generalization impede the development of standardized LLM-based solutions in dermatology.
Objective:
This review seeks to systematically evaluate the recent advancements, limitations, and future directions of applying LLMs in dermatology, with the goal of identifying key research priorities and informing the development of AI-assisted dermatological care. Specifically, we seek to: (1) summarize the state-of-the-art performance of LLMs across various dermatological tasks; (2) compare the technical characteristics and applicability of different LLM architectures; (3) analyze the data-related, methodological, ethical, and practical challenges hindering their widespread adoption; (4) propose actionable recommendations for future research and development to fully harness the potential of LLMs in revolutionizing dermatological practices.
Methods:
A systematic literature search was conducted in PubMed, Web of Science, and other databases for studies published between January 2018 and May 2024 that applied LLMs in dermatology and reported quantitative or qualitative results. Two reviewers independently screened the literature, extracted data, and assessed quality following PRISMA guidelines.
Results:
A systematic literature search was conducted in PubMed, Web of Science, and other databases for studies published between January 2018 and May 2024 that applied LLMs in dermatology and reported quantitative or qualitative results. Two reviewers independently screened the literature, extracted data, and assessed quality following PRISMA guidelines.
Conclusions:
LLMs hold significant potential for advancing dermatological diagnosis and treatment, but key challenges remain. Future research should focus on improving model generalization through multimodal data fusion and comprehensive knowledge integration while developing collaborative diagnostic frameworks. Multidisciplinary efforts in data governance, ethics, and regulation are crucial to ensure the safe, effective, and equitable implementation of these technologies.
Citation
