The US Food and Drug Administration has granted 510(k) clearance to Eyenuk Inc., a global artificial intelligence medical technology and services company to market its EyeArt® autonomous AI System for detecting Diabetes-related Retinopathy1.
Diabetes-related Retinopathy is the most common form of Diabetes-related eye disease, affecting approximately 93 million people worldwide and existing in two main forms; non-proliferative and proliferative Diabetes-related Retinopathy. Non-proliferative Diabetes-related Retinopathy is the early form in which high glucose levels in the bloodstream causes blockages to develop in blood-vessels within the eye, resulting in some leakage. In proliferative Diabetes-related Retinopathy, new and immature blood-vessels grow in an effort to compensate for the damaged blood-vessels. However, these newly-formed blood-vessels are weaker and burst, leaking blood into the vitreous cavity, causing “floaters” or “spots” to develop in the visual field.
The EyeArt® system is designed to automatically detect more than mild Diabetes-related Retinopathy and vision-threatening Diabetes-related Retinopathy in adults living with Diabetes with a single test. Clearance was based on results from a prospective clinical trial involving 380 individuals at seven primary care sites and 275 individuals at four ophthalmology sites. The EyeArt® system demonstrated 96% sensitivity and 88% specificity in detecting more than minor Diabetes-related Retinopathy and 92% sensitivity and 94% specificity in detecting vision-threatening Diabetes-related Retinopathy2.
The system is indicated for use with the Canon CR-2 AF and the Canon CR-2 Plus AF fundus cameras and Eyenuk are actively planning to expand the list of Eyeart-compatible imaging devices. Additionally, Autonomous AI for diabetic retinopathy screening will likely be covered by payors starting 2021.
References: