Using artificial intelligence to spot the signs of diabetic eye disease

In summary

Diabetic retinopathy is a leading cause of blindness in adults with diabetes. With over 2 million eye screening appointments annually in England, current systems are under pressure.

City St George’s researchers are evaluating AI-based tools to support diagnosis and improve screening efficiency.

What did we explore and how?

Professor Alicja Rudnicka and her team examined whether artificial intelligence (AI) software could support national diabetic eye screening programmes by identifying high-risk images of diabetic retinopathy.

In a study funded by the National Institute for Health Research, they tested the AI system on 20,000 patients. The goal was to see if AI could triage images and reduce the burden on human graders.

The AI performed particularly well in detecting sight-threatening diabetic retinopathy. By pre-screening images, it could cut the workload of human graders by 50%.

We undertook a study examining the accuracy of AI software to detect diabetic retinopathy, to see whether we could partially replace humans with the software to screen for diabetic eye disease.

Benefits and influence of this research

If adopted nationally, the AI software could save the NHS over £10 million per year by reducing manual grading demands and improving early diagnosis rates.

Rudnicka and Professor Christopher Owen are now working with Moorfields Eye Hospital and Homerton University Hospital to test the AI system in real-world settings, with the goal of influencing clinical practice across the NHS.

We hope that the system will soon be introduced in one of our partner centres in London and that it will be rolled out across other diabetic eye screening centres within the next five years, potentially saving the NHS over £10 million per year.