New AI system detects retinopathy with 95.5% accuracy

A new artificial intelligence system is capable of detecting diabetic retinopathy 95.5% of the time. The technology was described at the annual meeting of the American Academy of Ophthalmology (12–15 October, San Francisco). The system, named EyeArt, was used to screen 893 patients with diabetes in 15 different medical locations. It can provide a reading within 60 seconds. EyeArt displayed 95.5% sensitivity and 86% specificity, while more than 90% of eyes flagged as positive by the system had diabetic retinopathy or another eye disease.