Artificial Intelligence for Eye Health: From Innovation to Implementation and Global Impact

Vision impairment and avoidable blindness remain major global public health challenges, and disproportionately affect low- and middle-income countries and underserved populations. Advances in artificial intelligence (AI), particularly deep learning for retinal image analysis, now allow automated detection of diabetic retinopathy, glaucoma, age-related macular degeneration and retinopathy of prematurity with accuracy comparable to that of expert graders. Smartphone-based fundus imaging, teleophthalmology and large-scale screening platforms are extending eye care beyond tertiary hospitals. However, implementation remains limited by algorithmic bias, a lack of external validation, gaps in regulation and data governance, uneven digital infrastructure and weak integration into existing health systems. AI in eye health now needs to move beyond technical proof of concept towards equitable, scalable and sustainable public health impact.This Research Topic aims to advance the translation of AI-driven innovations into effective and equitable public health practice for eye health. While AI algorithms for ocular imaging have matured rapidly, evidence on real-world utility, cost-effectiveness and population-level outcomes remains fragmented. By bringing together interdisciplinary evidence from algorithm development, implementation science, health economics, policy and ethics, this collection seeks to consolidate lessons learned and inform evidence-based strategies that use AI to reduce avoidable vision loss and strengthen eye health systems worldwide.We welcome studies that address a clear population-level public health question, such as scalable screening models, real-world evaluations of AI tools in community settings, and governance frameworks for responsible data use. Studies limited to algorithm development or diagnostic accuracy, without a public health application, fall outside the scope of this Research Topic.Suitable themes for manuscripts include, but are not limited to:• AI-enabled population screening for major causes of avoidable vision loss, such as diabetic retinopathy, glaucoma, age-related macular degeneration, retinopathy of prematurity and cataract, including teleophthalmology, smartphone imaging and store-and-forward models;• external validation, generalisability and algorithmic fairness across multi-ethnic populations, imaging devices and settings;• implementation and health system integration, including workflow design, task-shifting, deployment in primary care and communities, human–AI collaboration and clinician trust;• health economics, including cost per case detected, budget impact and resource allocation in low-resource settings;• ethics, regulation and governance, including data privacy, informed consent, liability and equitable data sharing;• population-level data and public health surveillance, including AI for epidemiological mapping, risk prediction and priority-setting;• equity, community perspectives and capacity building, including acceptability, digital health literacy, workforce development and South–South collaboration. Topic editors: Jie Hao, Wenyi Hu Source: https://www.frontiersin.org/research-topics/87170/artificial-intelligence-for-eye-health-from-innovation-to-implementation-and-global-impact