Researchers at King's College London have developed an AI model capable of predicting the risk of sight-threatening diabetic retinopathy (DR) up to three years in advance using anonymized NHS eye data from over 100,000 diabetes patients.
Diabetic retinopathy affects approximately one in three individuals with diabetes and is a leading cause of vision loss among working-age adults. Currently, the NHS Diabetic Eye Screening Program (DESP) screens around 3.2 million people annually, costing over £85 million in England alone.
The new AI model can enable personalized screening by assessing whether individuals are at low or high risk of developing DR within one, two, or three years, utilizing retinal images—a feature not presently available in NHS screenings.
To create the model, researchers analyzed more than 1.2 million retinal images from the South East London DESP and validated it with around 70,000 images from the INSIGHT Health Data Research Hub.
Professor Timothy Jackson highlighted the importance of using routinely collected clinical data for patient benefit, while Professor Christos Bergeles noted that AI could modernize diabetic screening without compromising its effectiveness in preventing vision loss.
If implemented, this individualized approach could significantly reduce the screening burden for low-risk individuals and ensure timely care for those at high risk, potentially saving the NHS millions and thousands of appointments each year.