AI breakthrough improves eye exam rates in youth with diabetes
In a ground-breaking study conducted by the Johns Hopkins Children’s Center, researchers have demonstrated a significant leap forward in the field of diabetic eye care through the implementation of autonomous artificial intelligence (AI) for eye examinations. This innovative approach has markedly improved the rate of screening completions among children and adolescents with diabetes, thereby offering a promising avenue to combat the incidence of diabetes-related eye diseases (DED), which can lead to blindness if left unchecked.
The study, which has been a focal point of attention in a recent Johns Hopkins news release, utilises a non-intrusive method whereby images of the retina are captured without the necessity for pupil dilation. Following this, AI technology analyses the images to provide instantaneous results, distinguishing this method from traditional screening procedures which often require an additional appointment and dilation of the eyes.
The significance of this research cannot be overstated, especially considering its potential to bridge healthcare disparities. Historically, minority and economically disadvantaged youth, who are at a heightened risk of developing DED, have faced substantial barriers in accessing regular eye screenings. The AI-driven method heralded by this study not only promises to close these care gaps but also to enhance adherence to screening protocols among these populations.
Published in the esteemed journal Nature Communications, the study analysed eye exam completion rates among individuals under 21 years of age suffering from type 1 and type 2 diabetes. Remarkably, it was found that all participants who underwent the AI-based screenings completed their eye assessments, a stark contrast to traditionally lower adherence rates.
Diabetic retinopathy, a condition affecting 4% to 9% of youth with type 1 diabetes and 4% to 15% of those with type 2 diabetes, underscores the urgency for regular screenings. The American Diabetes Association estimates that approximately 238,000 individuals under the age of 20 are living with diagnosed diabetes, making early detection and treatment of eye conditions critical to preventing the advancement of DED.
Despite the general recommendation for annual screenings, traditional methods have seen a participation rate of only 35% to 72% among young diabetic patients, with even lower rates observed in minority and economically disadvantaged groups. Barriers such as confusion over the necessity of screenings, inconvenience, and lack of accessibility have all contributed to this shortfall.
The Johns Hopkins study introduced a novel solution to these challenges by incorporating autonomous AI screening into routine visits to the endocrinologist, thereby eliminating the need for separate appointments and the discomfort of eye dilation. This method, which involves taking four images of the eye to assess for diabetic retinopathy, has not only streamlined the screening process but also facilitated immediate follow-up actions when necessary.
This research initiative enrolled 164 participants from the Johns Hopkins Pediatric Diabetes Center, with a demographic makeup that was both gender and ethnically diverse. The findings revealed a 100% completion rate for eye exams among the group subjected to AI screenings, a significant improvement over traditional methods.
The study’s lead, Dr. Risa Wolf, emphasised the dual benefits of this approach: increased screening rates and the potential to enhance health equity. By making screenings more accessible and convenient, the researchers hope to prevent the progression of diabetic eye disease across all demographics.
However, it is important to note the study’s limitations, including the current FDA approval status of the autonomous AI for individuals under 21 and the potential bias due to some participants’ prior familiarity with AI screenings from a previous study.
Funded by the National Eye Institute of the National Institutes of Health and the Diabetes Research Connection, this study represents a pivotal step forward in diabetic eye care. It not only underscores the transformative potential of AI in healthcare but also highlights the critical need for innovative solutions to improve access and outcomes for vulnerable populations.