AI models used to automate remote screening for diabetic retinopathy
The algorithms could provide major accessibility improvements and cost savings to underdeveloped regions
The new AI-enhanced screening tool could be pivotal to treating diabetic retinopathy in rural areas. Image credit: ©bannafarsai – stock.adobe.com
A collaborative research team announced the development of artificial intelligence (AI) algorithms to automate remote ocular screenings for diabetic retinopathy (DR). A news release1 announcing the automated machine learning models (AutoML) attributed their creation to investigators from Queen’s University Belfast, Northern Ireland; The Philippine Eye Research Institute, Manila, Philippines; and Joslin Diabetes Center of Harvard Medical School, Boston, Massachusetts, US.
According to the news release, DR is the leading cause of blindness globally among working-age adults. Currently, 537 million adults live with diabetes, and that number is projected to rise to 643 million adults by 2030, subsequently increasing the prevalence of DR. Deploying widescale, systematic DR screening programmes (DRSP) is a challenge, particularly in countries with fewer resources. For example, in the Philippines, population density and lack of healthcare infrastructure are major obstacles. Geographic isolation also plays a part: the country comprises 7,641 islands.
Thus, the AI model used in this study works in conjunction with remote imaging to make screenings accessible. A portable, handheld retina camera is used to photograph the retina, and these images are then transmitted to a centralised database for assessment. The results, including follow-up instructions, are made available to the patient online and at the screening sites. The AutoML developed by the collaborative research team allows more images to be assessed accurately, and at less cost, than using a team solely composed of human graders. By combining community-based screening teams and AI integration, this technology can provide specialised care to at-risk populations.
Professor Tunde Peto, professor of clinical ophthalmology from the School of Medicine, Dentistry and Biomedical Sciences at Queen’s University Belfast, said the technology can be scaled up to save sight for drastically underserved populations.1
“AutoML allows the development of code-free algorithms at minimal cost by individuals without extensive background in computer programming language,” Professor Peto explained. “As coding skills are not common among healthcare workers, the use of AutoML models for DR screening can potentially address disparities in the delivery of eye care to patients with diabetic eye disease by ensuring a more rapid assessment of retinal images at point of care with minimal cost, thereby ensuring prompt referrals and timely intervention for those patients who require more specialised eye care.
Reference
1. AI technology developed by Queen’s University as part of global research project could help tackle leading cause of blindness. Queen's University, Belfast. News release. Accessed 28 February, 2024.
