Verily (formerly Google Life Sciences) and Nikon’s retinal imaging unit, Optos, are teaming up to build machine learning algorithms that can detect signs of retinopathy in patients with diabetes.
These tools are useful in helping doctors prioritize the patients that are most urgently in need of care. Diabetic retinopathy affects 40% to 45% of people with diabetes. Without treatment, it often leads to severe vision loss.
Verily isn’t the first technology company to develop new tools to detect the disease. Last year, data science startup Kaggle launched a competition sponsored by the California Healthcare Foundation to develop a sophisticated algorithm that can read medical images of the retina. Read more about Verily’s algorithm in JAMA.