CorneaAI, a deep learning artificial intelligence model created to diagnose cataracts and corneal diseases was shown to enhance the diagnostic accuracy of ophthalmologists, according to a study in Nature.1
Specifically, the overall diagnostic accuracy of the ophthalmologists increased from 79.2% to 88.8% when using CorneaAI.
The study was comprised of 40 ophthalmologists (20 specialists and 20 residents) who classified 50 iPhone 13 Pro photos and 50 diffuser slit-lamp photos into nine corneal condition categories. These categories: (1) normal condition, (2) infectious keratitis, (3) immunological keratitis, (4) corneal scar, (5) corneal deposit, (6) bullous keratopathy, (7) ocular surface tumor, (8) cataract/intraocular lens opacity, and (9) primary angle-closure glaucoma). Both types of photos represented the same cases.
Read the full study here.
1. Maehara H, Ueno Y, Yamaguchi T, et al. Artificial intelligence support improves diagnosis accuracy in anterior segment eye diseases.Sci Rep. 2025;15(1):5117. Published 2025 Feb 11. doi:10.1038/s41598-025-89768-6