Google Found a Way to Detect Diabetic Eye Disease with Machine Learning


A Google research team began to study whether machine learning could help to detect Diabetic Retinopathy (an eye condition that affects people with diabetes). The research team was able to discover a deep learning algorithm capable of diagnosing signs of Diabetic Retinopathy using retinal photographs. This innovation can help doctors to predict Diabetic Retinopathy at earlier stage, which can prevent loss of vision of the patient.

The Google Research team said on a blog post "Working with a team of doctors in India and the U.S., we created a dataset of 128,000 images and used them to train a deep neural network to detect diabetic retinopathy. We then compared our algorithm’s performance to another set of images examined by a panel of board-certified ophthalmologists."

Now DeepMind team is working on applying machine learning to this algorithm. In the near future this method will assist doctors in diagnose more patients quickly and help them to prevent loss of vision.