Study finds AI-powered eye exams increase screening rates among young people with diabetes

A Johns Hopkins Children’s Center study of children and youth with diabetes concludes that so-called autonomous artificial intelligence (AI) diabetic eye exams significantly increase completion rates of screenings designed to prevent diabetic eye disease ( DED) potentially blinding. During the exam, photos are taken of the back of the eyes without dilating them, and AI is used to provide an immediate result.

The study noted that AI-based technology used in the exams could address “care gaps” among racial and ethnic minority youth with diabetes, populations with historically higher rates of DED and less access or adherence to regular screening for eye damage.

In a report on the study published in Natural communicationsinvestigators examined diabetic eye exam rates among people under 21 years of age with type 1 and type 2 diabetes and found that 100% of patients who underwent AI exams completed the vision assessment.

DED primarily refers to diabetic retinopathy, a potentially blinding complication of diabetes that occurs when poorly controlled sugar levels cause overgrowth or damage to blood vessels and nerve tissue in the light-sensitive retina, rear of the eye.

According to the study researchers, retinopathy affects between 4% and 9% of young people with type 1 diabetes and between 4% and 15% of young people with type 2 diabetes. It is estimated that approximately 238,000 children, Teenagers and young adults under the age of 20 have been diagnosed with diabetes, according to the American Diabetes Association. Frequent screenings for DED facilitate early detection and treatment and can help prevent progression of DED.

In general, diabetes specialists and ophthalmologists recommend annual screenings, which usually require an additional, separate visit to an eye care professional, such as an optometrist or ophthalmologist, and the use of drops to dilate the pupil to obtain clear vision of the retina. visible using specialized instruments.

However, studies show that only 35% to 72% of youth with diabetes receive recommended screenings, with even higher rates of care gap among minority and poor youth. Previous studies also show that barriers to screenings include confusion about the need for screenings, inconvenience, and lack of time, access to specialists, and transportation.

Previous studies by Risa Wolf, MD, a pediatric endocrinologist at Johns Hopkins Children’s Center, and her team found that autonomous AI screening using cameras produced results that allowed for an accurate diagnosis of DED.

In the new study, researchers recruited 164 participants, ages 8 to 21, all from the Johns Hopkins Pediatric Diabetes Center between November 24, 2021 and June 6, 2022. Some 58% were female and 41% were from minority groups (35% black; 6% Hispanic).

Some 47% of participants had Medicaid insurance. Subjects were randomly assigned to one of two groups. A group of 83 patients received standard screening instructions and care and were referred to an optometrist or ophthalmologist for an eye exam.

A second group of 81 patients underwent a five- to 10-minute standalone diabetic eye exam with an AI system during a visit to their endocrinologist (the specialists who typically treat people with diabetes) and received their results during the same visit.

The AI ​​system takes four photos of the eye without dilation and runs the images through an algorithm that determines the presence or absence of diabetic retinopathy, Wolf says. If present, a referral is made to an ophthalmologist for further evaluation. If he’s absent, “you’re good for the year, and you’ve just saved yourself some time,” she adds.

The researchers found that 100% of patients in the AI ​​stand-alone screening group completed their eye exam that day, while 22% of patients in the second group completed an eye exam within six month with an optometrist or ophthalmologist. The researchers found no statistical difference based on race, gender, or socioeconomic status in whether participants in the second group scheduled a separate screening with an ophthalmologist.

The researchers also found that 25 of 81 participants, or 31 percent, in the autonomous AI group had a result indicating the presence of DED. Sixteen of these participants, or 64%, then made an appointment with an eye care professional. Further analysis showed that those who did not make an appointment were more likely to be black and have Medicaid insurance.

“With AI technology, more people can be screened, which could then help identify more people who need follow-up evaluation,” Wolf says. “If we can offer this more easily at the point of care with their diabetes doctor, then we can potentially improve health equity and prevent the progression of diabetic eye disease.”

The investigators caution that the autonomous AI used in their study is not approved by the U.S. Food and Drug Administration for anyone under 21 years old. And they say that a potential source of bias in the study was that some of the participants were familiar with standalone AI diabetic eye exams from a previous study, and so they might have been more willing to participate in the news.