FYI.

This story is over 5 years old.

Tech

Google Looks to Detect Diabetic Eye Disease With Machine Learning

Google said its algorithm diagnosed diabetic retinopathy just as accurately as opthamologists.
Diabetic retinopathy. Image: Google

Google wants to help cure a common form of blindness with machine learning.

Diabetic retinopathy, a fast-growing cause of blindness worldwide, has the potential to affect about 415 million people with diabetes. If the disease is caught early, it can be treated before a patient begins to irrevocably lose their sight. The Centers for Disease Control and Prevention predicts 16 million people over the age of 40 will be affected with the disease by 2050.

Advertisement

This week, Google announced its foray into medical screening by using machine learning to improve the screening process for signs of diabetic retinopathy in diabetic patients. Researchers developed an algorithm that uses images of a patient's eye to diagnose the disease as consistently as a real opthamologist, according to a Google release.

"These are exciting results, but there is still a lot of work to do," Google produce manager Dr. Lily Peng and research engineer Varun Gulshan wrote in a company post.

Google said it plans to incorporate this technology with 3D imaging to analyze the layers of the eye in detail for better diagnosis. "We hope our study will be just one of many compelling examples to come demonstrating the ability of machine learning to help solve important problems in medical imaging in healthcare more broadly," the release said.

Meanwhile, other Silicon Valley technology staples have also ventured into medical science. Facebook CEO Mark Zuckerberg announced this year he and his wife would donate $3 billion to their private foundation to one day "cure all diseases."

Get six of our favorite Motherboard stories every day by signing up for our newsletter.