Artificial Intelligence Screenings Improve Access to Eye Care in Sub-Saharan Africa

ByMartha R. Camara

Jun 14, 2022

Orbis International has unveiled new research findings demonstrating that artificial intelligence (AI)-powered diabetic retinopathy screenings have improved patient timeliness and referral service utilization in Rwanda, which may help reduce vision loss due to diabetes in the area.

According to Orbis, the study is the first in the world to explore the use and direction of AI for diabetic retinopathy and also includes results that are broadly applicable in other settings for retinopathy screenings. diabetics assisted by AI.

According to Ciku Mathenge, MD, PhD, medical adviser for Orbis International and principal investigator of the study, the number of people with diabetes is increasing rapidly worldwide, with the largest projected increase in Africa, estimated at 143% by 2045.

“Unfortunately, screening programs for diabetes-related vision loss are often difficult to implement effectively in low-resource settings,” Mathenge said in a prepared statement. “Our research results prove that integrating advanced technologies such as artificial intelligence into diabetic retinopathy screenings is not only feasible, but also an effective solution to identify those at risk and improve access to care. for people with the disease, even in the most vulnerable communities.”

The study investigated whether using Orbis’ Cybersight AI tool to detect diabetic retinopathy in images taken of the eye during routine screenings would lead to increased patient use of diagnostic services. benchmark for diabetic retinopathy.

The screenings were carried out in partnership with the Rwandan International Institute of Ophthalmology (RIIO) at four diabetes clinics in and around Kigali, Rwanda in 2021. The participants, who were recruited during routine visits to the diabetologist, had been diagnosed with type 1 or 2 diabetes and were 18 years of age or older.

Research found that patients who received immediate feedback about their condition based on AI-supported screenings were more likely to attend recommended follow-up appointments than those who were only referred afterward. the completion of a human grading report, a process that can take several days. . Potential reasons for this have been attributed to a few factors:

  • Immediate feedback offers eye care teams the opportunity to educate a patient about the importance of follow-up care right after they become aware of their condition.
  • Immediate feedback also offers patients who need referrals the opportunity to visit the ophthalmologist the same day, minimizing the need for extra travel for an appointment.
  • Receiving an instant report including images of the retina – a visual reminder of its condition – may also have supported acceptance of recommended referral, particularly for asymptomatic patients.

The growing burden of diabetes and its associated complications is increasing pressure on healthcare systems around the world. Diabetic retinopathy, a complication of diabetes that affects the eyes, is caused by damage to the retinal blood vessels at the back of the eye. Poorly controlled blood sugar is a risk factor. If left untreated, diabetic retinopathy can lead to progressive loss of vision or irreversible blindness. It is estimated that by 2040, 224 million people worldwide will have some form of diabetic retinopathy, with vision at risk in 70 million of these people worldwide.

Early diagnosis and treatment of diabetic retinopathy through screening reduces vision loss by 98%. But low-resource countries like Rwanda often lack sufficient health care infrastructure and trained medical personnel to effectively implement diabetic retinopathy screening programs. Additionally, many patients who are referred for additional follow-up appointments for diabetic retinopathy often do not do so due to financial barriers, travel time, lack of clarity in the referral process, and lack of clarity. uncertainty about the treatability of the disease.

Recent advances in AI offer a promising opportunity to scale effective and successful diabetic retinopathy programs, especially in areas where there is a shortage of ophthalmologists, as other skilled healthcare personnel, such as nurses, can also be trained to use technology effectively. This dramatically increases the number of people who can be screened, helps identify people with diabetes at risk of blindness much faster, and reduces unnecessary referrals.

Cybersight, Orbis’ award-winning telemedicine platform, provides eyecare professionals in areas that need it most free virtual access to training and other resources to better support their patients. One of the biggest assets of the platform is its artificial intelligence tool, Cybersight AI. Cybersight AI can detect abnormalities often associated with common eye diseases – like glaucoma, diabetic retinopathy and macular disease – in just seconds by analyzing images of the back of the eye taken during routine exams . When diseases like these are detected and treated early, patients have the best chance of not losing their sight.

The research was funded in part through Mathenge’s receipt of a Roche Collaborative Research Fellowship from the Association for Research and Vision in Ophthalmology (ARVO).

According to the press release, the fellowship pairs early career researchers from developing countries with collaborating scientists in well-established research laboratories with the aim of building research capacity in ophthalmology worldwide.