<![CDATA[Glaucoma is a leading cause of irreversible blindness worldwide, with many cases remaining undiagnosed due to limitations in traditional screening methods. Conventional diagnostic approaches rely on specialized equipment, trained clinicians, and subjective interpretation, restricting large-scale and early detection, particularly in resource-limited settings. Artificial intelligence (AI)-based screening methods have emerged as scalable and objective solutions for automated glaucoma detection using retinal imaging data. This review provides a comprehensive overview of recent advances in AI-driven glaucoma screening, focusing on methodological innovations, diagnostic performance, fairness considerations, and real-world implementation challenges. A systematic analysis of studies published up to early 2025 was conducted, covering AI applications in fundus photography, optical coherence tomography (OCT), and multimodal imaging. Approaches including deep learning-based classification, segmentation, and progression prediction are evaluated. Recent AI models demonstrate high diagnostic performance, with reported accuracies of 95–98% and strong sensitivity and specificity. Multimodal fusion enhances early detection and progression monitoring, while explainable AI techniques improve transparency by highlighting clinically relevant retinal regions. Fairness-aware strategies further address demographic disparities to support equitable screening. Lightweight architectures enable portable and mobile deployment for large-scale community screening. AI significantly improves the accuracy, accessibility, and scalability of glaucoma detection. Continued emphasis on data diversity, interpretability, and clinical validation is essential for sustainable integration into real-world ophthalmic practice.]]>
