Diabetic eye disease, formally known as Diabetic Retinopathy, is a complication resulting from diabetes that affects the eyes. Caused by damage to the blood vessels of the retina—the light-sensitive tissue found at the back of the eye—diabetic retinopathy can ultimately lead to irreversible blindness. Statistics reveal that the disease will affect up to 80% of those who have had diabetes for over two decades.
While proper treatment protocols and consistent monitoring of the eyes can substantially reduce the disease’s occurrence, most medical specialists capable of early detection are not obtainable in parts of the world in which the disease is prevalent. The examination is costly, requiring highly trained specialists to interpret images and rate them for disease presence and severity, and inaccessible for most parts of the underdeveloped world.
Researchers at Google, however, have recently developed a deep learning algorithm that is capable of interpreting the signs and symptoms of diabetic eye disease in retinal photographs: potentially assisting physicians in screening more patients in settings with limited resources. These new advances have the potential to solve a host of critically important medical and healthcare problems; soon, automated screening methods for diabetic eye disease may be available to assist doctors in evaluating greater numbers of patients, and rapidly refer those with signs of disease to specialists.