The word AI or Artificial Intelligence gets thrown around a bit too much these days, especially with the intent of describing it as something dangerous. We’ve seen numerous uses of Artificial Intelligence or AI, which range from creating a virtual TV news anchor to finding new drugs for better health. While some people advocate against the use of AI fearing that it could take over the world, that scenario currently seems to be a far cry from reality.
Researchers have now found a new way to employ AI to our advantage; to screen people for Diabetic Eye Disease. A new research has proved that an AI screening system can accurately detect diabetic retinopathy 95.5 percent of the time, and that too in merely 60 seconds. This means that a non-ophthalmologist can now use the system and help people discover the eye disease which, if untreated, can potentially lead to blindness.
EyeArt is an Artificially Intelligent system that is capable of accurately detecting diabetic retinopathy in patents 95.5 percent of the time. It has been tested in the real-world on more than half a million patients and nearly two million retinal images globally. It takes about 60 seconds to run a test and is touted to help any physician detect referable diabetic retinopathy patients during a diabetic patient’s regular examination.
Tried and tested
It can be difficult to solely rely on an AI based software system and it needed to be validated before being deployed widely. The MD of the Doheny Eye Institute/UCLA, Srinivas Sadda, along with his colleagues across the US compared EyeArt against experts using the gold-standard grading system (ETDRS) as part of a trial. The system was used at 15 different medical locations to screen 893 diabetic patients.
The results generated by it were reviewed for clinical accuracy by certified graders. As per EyeArt, their system’s sensitivity was 95.5 percent, and specificity was 86 percent while using only undilated images (patients’ pupils were not dilated). Do note that higher sensitivity means safety and high specificity is related with effectiveness.
“Accurate, realtime diagnosis holds great promise for the millions of patients living with diabetes. In addition to increased accessibility, a prompt diagnosis made possible with AI means identifying those at risk of blindness and getting them in front of an ophthalmologist for treatment before it is too late,” says Srinivas Sadda, MD of the Doheny Eye Institute/UCLA.
Image Credits: Shutterstock
Stay tuned to Silicon Canals for more European technology news.