ARTIFICIAL INTELLIGENCE IN OPHTHALMOLOGY: TRANSFORMING DIAGNOSIS AND PATIENT CARE

Background: Artificial intelligence is rapidly transforming ophthalmology by improving diagnostic accuracy, clinical efficiency, disease monitoring, and access to eye care. Machine learning and deep learning models are increasingly being applied to ophthalmic imaging modalities such as fundus photography and optical coherence tomography for detecting diabetic retinopathy, glaucoma, age-related macular degeneration, and other retinal disorders. Objective: To review the current progress, clinical applications, limitations, and future perspectives of artificial intelligence in ophthalmology, with emphasis on diagnosis, screening, teleophthalmology, and patient-centered care. Methods: A literature review was conducted using databases including PubMed, Scopus, Web of Science, and IEEE Xplore. Relevant peer-reviewed studies focusing on artificial intelligence, machine learning, deep learning, ophthalmology, diabetic retinopathy, glaucoma, age-related macular degeneration, optical coherence tomography, retinal imaging, and teleophthalmology were reviewed. Studies reporting diagnostic performance, clinical application, predictive value, or implementation challenges were included. Results: Artificial intelligence has demonstrated strong diagnostic performance in image-based detection of diabetic retinopathy, glaucoma, and age-related macular degeneration. Deep learning systems have shown high accuracy in analyzing fundus photographs and OCT scans, often achieving performance comparable to expert clinicians. AI also supports teleophthalmology, automated screening, disease progression monitoring, referral decision-making, and risk prediction. However, several challenges remain, including dataset limitations, algorithmic bias, poor generalizability, lack of explainability, regulatory uncertainty, privacy concerns, limited workflow integration, and insufficient prospective validation. Conclusion: Artificial intelligence has considerable potential to improve ophthalmic diagnosis, screening, treatment planning, and access to eye care, particularly in underserved settings. Despite promising advances, successful clinical implementation requires diverse datasets, explainable models, ethical data governance, standardized validation, and integration into real-world ophthalmic workflows. Future research should focus on prospective, multicenter, and patient-centered studies to ensure safe, equitable, and effective use of AI in ophthalmology. Keywords: Artificial Intelligence; Ophthalmology; Deep Learning; Machine Learning; Retinal Disease Detection; Medical Imaging; Diabetic Retinopathy; Glaucoma; Age-Related Macular Degeneration; Teleophthalmology.