RETINAR AI platform developed to detect diabetic retinopathy

Researchers from CONICET and the National University of the Center of the Province of Buenos Aires (UNICEN) have developed RETINAR, an artificial intelligence platform designed for the early detection of diabetic retinopathy. The system analyzes fundus photographs to identify signs of the disease, classifying results on a scale of zero to five. If a high-risk grade is detected, the case is referred to an ophthalmologist for final clinical decision-making.

Tested in public hospitals in Buenos Aires province, the tool demonstrated high accuracy, recording only one false negative in a trial of approximately 400 patients. RETINAR has been integrated into the Buenos Aires province's Integrated Health History system. The developers are currently seeking final authorization from ANMAT for commercialization and aim to expand the technology's use across Latin America to assist regions facing shortages of eye specialists.