Montreal hospital uses AI screening tool to detect eye disease in diabetes patients
The Centre hospitalier de l’Université de Montréal (CHUM) recently unveiled an Artificial Intelligence diagnostic system to identify eye problems in people with diabetes that can lead to blindness. The pilot project is the first of its kind in Canada to offer this quick, safe and reliable exam to at-risk patients.
Diabetic retinopathy is a frequent complication of diabetes and the leading cause of blindness affecting 500,000 Canadians. The condition is caused by lesions to the small blood vessels and neurons of the retina, the light-sensitive tissue lining the back of the eye. It often develops and progresses without symptoms, and may go unnoticed until irreversible vision loss occurs.
With this new telemedicine platform, a non-invasive test will reveal any anomalies via digital posterior images of the patient’s eye taken with a special retinal camera. These will then be sorted by an algorithm used by co-collaborator, Diagnos Inc., a Quebec company renowned for its CARA Tele-Retinal technology in 16 countries. Depending on the level of eye damage, patients may be referred for follow-up as needed with an ophthalmologist.
The Diagnos AI system, which has been approved by Health Canada and the U.S. Food and Drug Administration, has the potential to maximize access to testing in a simple and efficient way. Nonetheless, CHUM doctors will also be scrutinizing the first 100 patients and comparing their results with the algorithms.
The cost of the pilot project is $400,000, with the Quebec government contributing $160,000.
This new technological advancement could provide an accelerated path to preventing retinopathy diagnoses, the most common vision complication for people with diabetes.
Visit our JDRF Blog for more informative articles on type 1 diabetes and health.