Can novel retinal imaging and AI transform diabetic eye screening?
Research type
Research Study
Full title
Can novel retinal imaging modalities and AI decision support systems revolutionise DR screening in the 21st Century?
IRAS ID
363797
Contact name
Julie-Anne Little
Contact email
Sponsor organisation
Ulster University
Duration of Study in the UK
1 years, 11 months, 13 days
Research summary
Diabetic retinopathy (DR) is one of the most common diabetes-related complications, and is the leading cause of blindness among the working age population. DR damages blood vessels at the back of the eye on the retina. DR often has no symptoms at its early stages, but if it is not detected and treated in time, it can lead to permanent sight loss. Regular eye screening is therefore essential for people with diabetes. Screening usually requires specialist staff, non-portable and expensive equipment, and the use of a dilating eye drop, which can be uncomfortable and require extra waiting time for patient. Hence, many countries struggle to provide effective DR screening services due to lack of infrastructure and human resource. The barriers are even higher in low-and middle-income countries where most people with diabetes live globally and where there are fewer eye care specialists.
With the recent advancements in technology, portable handheld cameras may help make screening more accessible, because they are cheaper and can be used outside hospital settings. Some of these cameras also include artificial intelligence (AI) tools, which can diagnose DR and reduce the requirement of eye specialists. Although previous studies show promise, image quality without dilation is often not good enough, and it is not yet known which patients can be successfully photographed without the need for dilation.
This study will explore how well a handheld retinal camera with integrated AI performs when photographs are taken without dilating the pupils of the eye. We will examine whether factors such as age, baseline pupil size, and diabetic duration and severity affect image quality. Understanding this will help determine when non-dilated imaging is feasible and could support the development of more accessible imaging for people with diabetes.REC name
London - Central Research Ethics Committee
REC reference
26/PR/0486
Date of REC Opinion
15 Jul 2026
REC opinion
Further Information Favourable Opinion