Artificial intelligence and machine learning – automatic retinal images reading to detect diabetic retinopathy and other eye diseases

Reading of retinal images is time consuming and still a largely manual process. As we scale up the Tele-retina screening program, significant barriers in the time dedicated by technicians and physicians will exist.  Although the hope is to screen all Canadians at risk of DR, the reading of images will be a major constraint in the timely turn-around of diagnostics and therapeutics.  As an exciting new initiative, Diabetes Action Canada is currently collaborating with a group of investigators at the University of Montreal – Department of Ophthalmology, the Montreal Polytechnique, and Montreal Institute for Learning Algorithms (MILA) to investigate the role of artificial intelligence (AI) in retinal image analytics. This group of scientists are developing new algorithms using advanced technology to read retinal fundus photos and optical coherence tomography (OCT) images for diagnosis of diabetic retinopathy and other eye diseases.  The AI initiative will advance the improvement of care and outcomes of those individuals living with diabetes, improving access to eye care specialists and increasing clinician productivity.


Felfeli; R. Alon; R. Merritt; M.H. Brent. Toronto tele-retinal screening program for the detection of diabetic retinopathy and macular edema in urban and rural settings: a case series. Can J Ophthalmol. 2018 available on-line 23 Aug 2018

Boucher MC, Desroches G, Garcia-Salinas R, Kherani A, Maberley D, Olivier S, et al. Teleophthalmology screening for diabetic retinopathy through mobile imaging units within Canada. Can J Ophthalmol. 2008;43(6):658-68.

Hooper P, Boucher MC, Cruess A, Dawson KG, Delpero W, Greve M, et al. Canadian Ophthalmological Society evidence-based clinical practice guidelines for the management of diabetic retinopathy. Can J Ophthalmol. 2012;47(2 Suppl):S1-30, S1-54.

Stay updated with News and Events