Multimodal artificial intelligence in ophthalmology: Applications, challenges, and future directions.

Abstract

With the rapid development of artificial intelligence (AI) technology, multimodal AI that integrates multiple data modalities has shown tremendous potential in the field of ophthalmology. We systematically evaluate the current applications, technical characteristics, and clinical value of multimodal AI in ophthalmology. This study was conducted as a systematic review following the PRISMA guidelines. We searched relevant literature published from 2018 to 2025 in PubMed, Web of Science, Scopus, and Google Scholar databases. A total of 10 studies were included in the final analysis. The main applications of multimodal AI in ophthalmology include glaucoma, age-related macular degeneration, corneal diseases, ophthalmic emergency triage, cognitive impairment screening, diabetes complication screening, and chatbot-based ophthalmic consultation. Multimodal systems showed superior performance compared to unimodal systems across various application areas, with AUC improvements of 4-5% and accuracy improvements of 2-7%. Multimodal AI demonstrates broad application prospects in ophthalmology, providing more comprehensive and accurate diagnostic information. Future research should focus on clinical validation, novel fusion methods, interpretability, and lightweight models to promote the clinical translation of multimodal AI in ophthalmology.