Orbis research shows the power of AI for improving productivity of eye teams

Published in Nature, the peer-reviewed study is the first to demonstrate significant improvements in the productivity of medical professionals when using autonomous artificial intelligence (AI) to test for diabetic eye diseases.

Orbis research shows that with the help of AI, healthcare professionals can see more patients in less time, without a reduction in the quality of care. In fact, AI was so useful in detecting diabetic retinopathy that it improved health care worker efficiency by an incredible 40%.

This is the first randomized trial ever to be published on the efficiency of autonomous AI in eye care.

The study, which took place over five months at Deep Eye Care Foundation’s retina specialist clinic in Bangladesh, included over 2,000 patients. While one group of patients moved on to see a specialist regardless of the results of the AI screening, other patients did not see a specialist if the autonomous AI did not detect signs of diabetic retinopathy.

The study shows that using autonomous AI (AI that has not been assisted by human intervention) to screen patients for diabetic retinopathy resulted in a 40% increase in completed eye exams per hour — without any impact on the quality of care provided.

With AI, doctors no longer had to spend unnecessary time seeing patients who did not require further care. AI correctly diagnosed these patients’ eyes as healthy so that the medical professionals could focus on the patients whose vision was threatened.

Eye care professionals using AI to screen patients found a range of benefits. They reported that AI allows them to focus their time on more complex cases that require specialized treatment, crucial hours they previously had to split. It also means patients with simpler cases can be in and out of the clinic more quickly as they no longer have to wait for a specialist. For these patients, quick appointments led to higher satisfaction, especially for those from areas of limited resources who cannot afford to take time away from work and home.


A Lasting Partnership

Speaking about the global state of blindness and how AI can alleviate some issues, Country Director of Orbis Bangladesh, Dr. Munir Ahmed, says: “An estimated one billion people worldwide lack access to essential health services. This disparity hinders global economic growth and perpetuates the widening poverty gap. One way to tackle this problem is to increase efficiency and productivity using technology. The autonomous AI does not take the place of a doctor; rather, it is a tool to better streamline the clinical process."

As Dr. Ahmed suggests, AI is not a direct replacement for personal medical care — but rather a tool to assist eye care professionals to improve patient outcomes.

A male patient is screened for diabetic retinopathy using AI in an eye clinic in Rwanda

A woman in Rwanda has her eye screened for diabetic retinopathy using AI.

Dr. Khairul Islam, Executive Director of Deep Eye Care Foundation, explains: “this study is a significant breakthrough in the field of eye care for the low- and middle-income countries where scarcity of skilled health professionals and facilities are always a hindrance. These challenges can be overcome through integrating autonomous AI, which increases the efficiency and productivity of our doctors and, thus, helps us reach more patients by providing them with timely eye care."

Blindness in Bangladesh

With only 1,200 ophthalmologists and a population of 164 million, tools like AI can effectively improve eye care in Bangladesh. AI allows doctors to work smarter and faster, which can be the difference between a life of impaired sight or clear vision.

Orbis has also been working with partners to tackle Bangladesh’s growing burden of diabetic retinopathy. Officially launched in 2019, the Retinopathy Referral Network was created to ensure people with diabetes do not go undiagnosed and to give them better access to the right kind of eye care. Within the referral network there are also vision units specifically designed to serve children with diabetes.

Since first touching down in Bangladesh in 1985, we’ve trained more than 36,000 medical professionals and screened millions of patients — the hope is that AI will help this number grow exponentially.

Thank You

We want to give a huge thank you to Digital Diagnostics and Deep Eye Care for sharing their technology and joining our mission to end avoidable blindness.

The critical insights we get from our research inform our evidence-based approach to tackling global vision loss. If you are interested in funding Orbis research, or our efforts to use technology to revolutionize eye care in low- and middle-income countries, please contact [email protected].

Help more doctors get access to sight-saving AI

Donate today!

Close the modal
Loading
Sorry there was an error.
Try again