Transforming Retinal Vascular Disease Classification: A Comprehensive Analysis of ChatGPT's Performance and Inference Abilities on Non-English Clinical Environment

ObjectiveTo evaluate the effectiveness and reasoning ability of ChatGPT in diagnosing retinal vascular diseases in the Chinese clinical environment. Materials and MethodsWe collected 1226 fundus fluorescein angiography reports and corresponding diagnosis written in Chinese, and tested ChatGPT with four prompting strategies (direct diagnosis or diagnosis with explanation and in Chinese or English). ResultsChatGPT using English prompt for direct diagnosis achieved the best performance, with F1-score of 80.05%, which was inferior to ophthalmologists (89.35%) but close to ophthalmologist interns (82.69%). Although ChatGPT can derive reasoning process with a low error rate, mistakes such as misinformation (1.96%), and hallucination (0.59%) still exist. Discussion and ConclusionsChatGPT can serve as a helpful medical assistant to provide diagnosis under non-English clinical environments, but there are still performance gaps, language disparity, and errors compared to professionals, which demonstrates the potential limitations and the desiration to continually explore more robust LLMs in ophthalmology practice.