Skip to content

Clinical research at UCL and Moorfields on AI tools to identify at-risk neonates of retinopathy of prematurity (ROP). 

Researchers at UCL and Moorfields Eye Hospital, UK, have reported clinical research using artificial intelligence (AI) tools to identify at-risk neonates of retinopathy of prematurity (ROP).  The study used a deep-learning AI model that may identify at-risk neonates and their AI approach proposes that the tool should improve access to screening in other regions that may have limited neonatal services and / or availability of trained of ophthalmologists.  The lead author of the study, Dr. Konstantinos Balaskas based in Moorfields Ophthalmic Reading Centre & Clinical AI Lab, said: “Retinopathy of prematurity is becoming increasingly common as survival rates of premature babies improve across the globe, and it is now the leading cause of childhood blindness in middle-income countries and in the US”.

Retinopathy of prematurity (ROP) is a proliferative retinal vascular disease, typically affecting preterm neonates with low-birthweight. In these neonates, abnormal blood vessels can leak or bleed, and possibly leading to retinal detachment and an estimated 50,000 children globally can become blind.  Researchers have stated that over half of neonatologists in a USA-based national survey reported a scarcity of available eye care specialists as a barrier to ROP screening. Telemedicine may provide a more efficient model for care delivery, allied with well-designed AI tools, could tackle many of these limitations.  The UK research team applied an ethnically diverse population in London, UK, and aimed to externally validate their AI tool ethnically, geographically, and socioeconomically in diverse populations in four countries (UK, Brazil, Egypt and UDA). Their approach used code-free deep learning, not reliant on the availability of expertly trained data scientists, thus being of particular potential benefit for low resource health-care settings.  The UK study used retinal images from 1,370 neonates admitted to a neonatal unit at Homerton University Hospital NHS Foundation Trust, London applying a cohort of babies who were either born at less than 32 weeks gestational age or had a birthweight of less than 1501g.

Following their results, the AI tool was found to be as effective as senior paediatric ophthalmologists in discriminating normal retinal images from those with ROP that could lead to blindness, detailed in their paper in the Lancet Digit Health 2023; 5: e340–49.  First author of the study, Dr. Siegfried Wagner (UCL Institute of Ophthalmology and Moorfields Eye Hospital) commented that: “our findings justify the continued investigation of AI tools to screen for ROP. We are now further validating our tool in multiple hospitals in the UK and are seeking to learn how people interact with the AI’s outputs, to understand how we could incorporate the tool into real world clinical settings. We hope that the tool will enable a trained nurse to take images that could be assessed by the AI tool, in order for a referral for treatment to be made without the need for an ophthalmologist to manually review the scans”.