Diabetes is considered a global epidemic, affecting 8.5% of the worldwide adult population[i] . By 2020, 10% of the Canadian population will be living with diabetes. Those with the disease have an elevated risk of eyesight deterioration, which is the leading cause of blindness in adults under 60 years of age (working age). Diabetic Retinopathy (DR) is a complication of diabetes and accounts for 80% of diabetic-related blindness. Early detection of DR by regular screening effectively avoids vision loss from diabetes as necessary treatments prevent irreversible retina damage.
Diabetes Action Canada recognizes DR screening as a top priority. This prevention measure is shared amongst provincially funded telehealth and health authorities that are establishing DR screening programs to reach under-served and vulnerable populations. Telehealth retinal screening programs have the potential to scale up as part of a broader population-based screening program. To accomplish this, diagnostic capacity constraints among Canadian ophthalmologists must be considered. To address this, Diabetes Action Canada is now collaborating with a group of investigators at the University of Montreal (U of M), Department of Ophthalmology and The Montreal Polytechnique, who collaborate with the Montreal Institute for Learning Algorithms (MILA). MILA is a federation of University of Montreal researchers focused on machine learning and artificial intelligence-based analytics. With funding from Diabetes Action Canada and the U of M, they plan to develop algorithms using advanced technology to read retinal fundus photo images and optical coherence tomography (OCT) images for diagnosis of diabetic retinopathy and other eye disease. The fundamental goal is to improve access to high quality ophthalmological care by reducing image reading times and increasing clinician productivity. Together, Diabetes Action Canada and U of M are interested in investigating the role of artificial intelligence (AI) in retinal image analytics and its potential for application in clinical contexts in Canada and beyond. Following algorithm development and validation, this group envisions creating a national consortium that could leverage advanced technology to deliver high-value retinal and OCT image analysis across Canada.
This research collaboration was recently featured in and article in Le Journal de Quebec found here.
[i] Global report on diabetes. World Health Organization. http://www.who.int/diabetes/global-report/en . Accesses Sept 25 2017.