Artificial Intelligence in Opthamology:A study on different AIML approaches for Glaucoma prediction

Abstract

Glaucoma is a leading cause of irreversible blindness, and artificial intelligence (AI) has emerged as a promising tool for its early detection and management. Recent studies span fundus- and OCT-based deep learning models, electronic health record–driven classifiers, and sensor-based systems incorporating ocular biomechanics and circadian signals. Meta-analyses confirm strong
diagnostic performance of image-based AI, yet highlight persistent challenges in progression prediction, generalizability, and interpretability. EHR- and sensor driven approaches provide complementary insights but remain limited by data quality and cohort size. This review synthesizes current advances, evaluates their limitations, and emphasizes the need for multimodal, explainable, and
externally validated AI frameworks to achieve robust and clinically translatable glaucoma prediction