Google's AI will determine diabetic retinopathy better than doctors

The rate of diagnosis of diabetic retinopathyvery important. Delays in this case are fraught with complete blindness. To facilitate the work is designed to use in one of the Indian clinics algorithm, which was developed by Google and Verily.

Diabetic retinopathy is a serious complication of diabetes mellitus that affects the eyeball vessels. It manifests itself in every ninth sick and can really cause blindness.

Staff at Aravind Ophthalmology Clinic,located in the city of Madurai, worked on the technology of automatic detection of this pathology for 16 years. But only two years ago, in collaboration with Google engineers, they managed to achieve acceptable speed and accuracy.

Algorithm trained to diagnose diabeticretinopathy on 128 thousand ophthalmic images of the eyeballs. The effectiveness of the method was tested on 2 sets of 12 thousand images each. The accuracy of the diagnosis reached 97.5 percent. And it is faster and more accurate than the doctors.

A similar study was conducted by Google atlast year in conjunction with the hospital Rajavithi in Thailand. The results of the algorithm were then similar: accuracy was 95 percent. And people-ophthalmologists were wrong in a quarter of cases.

Algorithm from Google recently received a sign of compliance with the standards of the European Union CE. Thanks to him, the doctors of the clinic managed to diagnose more patients with prescribing the necessary treatment.

Source: venturebeat