Comparative performance of large language models for patient-initiated ophthalmology consultations
ChatGPT-4o and DeepSeek-V3 demonstrated the highest overall performance and are best suited for laypersons seeking ophthalmic information. Doubao and Qwen, with their richer clinical terminology, better serve users with medical training, whereas Wenxin Yiyan 4.0 Turbo most effectively supports patients' pre-procedural understanding of diagnostic procedures. Prospective randomized controlled trials are required to determine whether integrating the top-performing model into pre-consultation triage improves patient comprehension.
Authors
- Yanwu Xu (ORCID: https://orcid.org/0000-0002-1779-931X)
- Wei Chi (ORCID: https://orcid.org/0000-0002-7424-3879)
- Weihua Yang (ORCID: https://orcid.org/0000-0002-7629-0193)
- Xiaoyan Wang (ORCID: https://orcid.org/0000-0002-7651-7730)
- Xinyu Cui (ORCID: https://orcid.org/0009-0004-5616-8387)
- Zilin Zhang (ORCID: https://orcid.org/0000-0003-2740-5043)
- Shiqi Zhou
- Mingxue Huang
Institutions
- Southwest Medical University (CN)
- Southern Medical University Shenzhen Hospital (CN)
- South China University of Technology (CN)
Publication Details
- Journal
- Frontiers in Public Health
- Published
- 2025-09-22
- DOI
- https://doi.org/10.3389/fpubh.2025.1673045
- Citations
- 5
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 3.27
Funders
- National Natural Science Foundation of China
- Sanming Project of Medicine in Shenzhen