Kent Peterson, senior editor, has also produced award-winning work in television and radio.
Medical breakthroughs often come with hefty price tags—but here’s one that doesn’t: a new tool to help prevent blindness in people with diabetes.
Stanford University researchers have developed an automated algorithm to detect and evaluate diabetic retinopathy, the most common cause of vision loss in people with diabetes. No expensive supercomputer is needed; it can be used on an ordinary personal computer or smartphone.
Thinking like a human
The new algorithm relies on artificial intelligence, using what’s called “deep learning” to enable a computer to mimic the human brain’s ability to make decisions. More than 75,000 images help the computer learn to recognize what different stages of diabetic retinopathy look like. Then it uses that knowledge to check other images for signs of the disease.
In a study published in Ophthalmology, researchers found the algorithm could correctly identify disease stages 94 percent of the time. Even experienced eye specialists may not always do much better.
The new technology won’t make your eye doctor obsolete, though. “What we showed is that an artificial intelligence-based grading algorithm can be used to identify, with high reliability, which patients should be referred to an ophthalmologist for further evaluation and treatment,” said Theodore Leng, MD, lead author of the study, in a news release.
Changing the world
More than 400 million people worldwide have diabetes, and in some places, there aren’t enough ophthalmologists to see them. Researchers say the new algorithm could reduce the global rate of diabetic vision loss by helping ophthalmologists be more efficient in reaching those who need care.
Get the right care
Don’t assume your eyes are okay if you haven’t noticed any problems. Nearly half of all Americans diagnosed with diabetes have some degree of diabetic retinopathy, but many don’t know it. In the early stages of the disease, symptoms usually aren’t noticed. Over time, it damages blood vessels in the retina, affecting eyesight. Early detection and proper treatment can reduce the risk of blindness by 95 percent, according to the National Eye Institute.
Although the new algorithm could run on your phone or laptop, it is not an app for self-care but a tool for professional use. Have a complete eye examination as often as your eye care provider recommends; frequency depends on your individual needs.
Have you had diabetic eye problems? Share your experiences and advice by commenting below.