Abstract: Glaucoma, the leading cause of irreversible blinding worldwide, is characterized by its subtle and progressive nature. Early detection and surveillance of its progression are crucial for effective management. The cup-to-disc ratio (CDR) is one of the important indicators in this regard.Artificial intelligence (AI), through deep learning algorithms, enables efficient image analysis. When combined with existing diagnostic methods, AI significantly improves the CDR calculation in glaucoma, while greatly reducing screening costs and the rate of missed diagnoses in the early stages. However, challenges remain, including limited model generalizability due to the high reuse of public datasets, dependence on image quality, limited availability of glaucoma disease databases for AI model development, the “black-box” nature of AI algorithms, and the lack of legal and regulatory frameworks related to AI. Nevertheless, AI offers substantial advantages in medical image interpretation and holds great promise for advancing glaucoma diagnosis and treatment toward greater precision and convenience.

Key words: Glaucoma, Artificial intelligence, Screening, Cup-to-disc ratio