Researchers at King’s College London have created an AI tool using NHS eye data from over 100,000 diabetic patients. This tool can predict, up to three years in advance, who is at high risk of developing diabetic retinopathy (DR), a condition that can cause vision loss.
The study was published in Communications Medicine. DR affects about 1 in 3 people with diabetes and is a major cause of vision loss in adults. In the UK, everyone with diabetes aged 12 and above is asked to attend yearly eye checks through the NHS Diabetic Eye Screening Program (DESP). This program screens 3.2 million people annually, costing over £85 million.
Currently, the NHS cannot predict which patients are at higher risk of developing severe DR. However, the new AI tool can analyze images of the back of the eye and predict if someone is at low or high risk of sight loss within one, two, or three years.
The AI model was developed using over 1.2 million retinal images from diabetic patients in South East London. It was then tested on another set of 70,000 images to make sure it works accurately for people from different backgrounds.
Professor Timothy Jackson said this research highlights the value of using clinical data to improve patient care. Professor Christos Bergeles added that AI could help modernize screening programs and reduce unnecessary screenings, saving the NHS millions of pounds each year.
Source: msn
