There could be good news on the horizon for individuals suffering from diabetes. A new algorithm may be able to help detect diabetic retinopathy—it could help save vision for many
Diabetic retinopathy is a progressive, eye disease that affects people with diabetes. Damage to the retina, which is light-sensing tissue located at the back of the eye, worsens over time and can lead to vision loss. As sources explain, “Over time, diabetes damages the blood vessels in the retina. Diabetic retinopathy occurs when these tiny blood vessels leak blood and other fluids. This causes the retinal tissue to swell, resulting in cloudy or blurred vision. The condition usually affects both eyes. The longer a person has diabetes, the more likely they will develop diabetic retinopathy. If left untreated, diabetic retinopathy can cause blindness.”
In the past, it was hard for health-care providers to predict who would be affected by diabetic retinopathy. However, Stanford University researchers are seeking to change this with a new, deep-learning algorithm that could revolutionize the face of diabetic-retinopathy detection. The new technology would serve as a priceless, diagnostic tool that would automate the process of identifying the degenerative, eye condition.
As study authors Rishab Gargeya and Theodore Leng explain, “The algorithm processed color-fundus images and classified them as healthy (no retinopathy) or having DR, identifying relevant cases for medical referral.”
Their algorithim’s diabetic-retinopathy-detection success is practically unmatched. In their study of its accuracy, they found the algorithm could detect the condition accurately—94 percent of the time. With such reliability, the team predicts universal implementation of their algorithm could dramatically reduce the number of people who lose their sight to diabetic retinopathy by hastening diagnosis and treatment.