Two Orbis studies show benefits of AI to rural eye care

A 2021 study published in the British Journal of Ophthalmology found that patient satisfaction rose to over 99% when they were screened for diabetic retinopathy using Orbis’s Cybersight  AI tool. The study also showed that more than 63% people preferred AI carrying out the tests rather than humans.

The screenings were carried out during routine appointments at four diabetes clinics in and around Kigali, Rwanda.

In partnership with the Rwanda International Institute of Ophthalmology, this was one of the first clinical studies on AI in diabetic retinopathy screening in Africa.

Factors that could have supported a high patient satisfaction include:

• People receiving their eye exam during their diabetes appointment. This saved time and the cost of travel, issues which often stops rural patients seeking treatment
• Most patients did not need to have their pupils dilated to carry out the AI screening, saving time and avoiding temporarily blurred vision
• Printed reports were available during the screening, which gave the patients the chance to learn about their condition immediately

Another Orbis study from 2022 published in Ophthalmology Science showed that diabetic retinopathy screenings supported by AI increased the speed of appointments. This meant that people were more likely to use these services.

Diabetes is on the rise globally, and diabetic retinopathy is the leading cause of vision loss among working-age people around the world affecting the livelihoods and futures of millions every year.

Anyone with diabetes is at risk of developing the condition, which can lead to irreversible blindness. The global burden of diabetic retinopathy is estimated at 93 million with recent reports showing that it is increasing in sub-Saharan Africa more than in any other region in the world.

Access to diabetic retinopathy screening is critical to preventing vision loss. Most patients with diabetes do not realise they have the disease until their vision is already affected irreversibly.

In sub-Saharan Africa, there is an average of just 3.7 ophthalmologists per 1 million people, compared to around 44 per million in the UK.

Training other medical staff to carry out diabetic retinopathy screenings is an effective way in low- and middle-income countries to preserve ophthalmologists’ time.

AI makes the screening process accessible for medical staff and patients alike in non-ophthalmic settings.

Orbis’s original research is vital to our work fighting avoidable blindness. What we learn is used to inform our evidence-based approach to building sustainable eye care systems so that everyone, everywhere can access the sight-saving care they need to thrive.

Regularly published in top-tier peer-reviewed journals, our research findings also informs the work of others in the eye care sector – improving outcomes for more and more people across the globe.

If you would like to find out more about how you can support Orbis’s vital research, please contact [email protected].