First AI device approved to detect diabetes-related eye problems

The software analyzes images of the eye taken with a retinal camera and offers one of two results: “more than mild diabetic retinopathy detected: refer to an eye care professional” or “negative for more than mild diabetic retinopathy; rescreen in 12 months.”


The FDA has approved marketing of the first medical device to use artificial intelligence to detect diabetes-related eye problems, the agency announced on April 11.

The software program, called IDx-DR, analyzes images of the eye taken with a retinal camera called the Topcon NW400, providing the physician with one of two results: “more than mild diabetic retinopathy detected: refer to an eye care professional” or “negative for more than mild diabetic retinopathy; rescreen in 12 months.”

In a study of 900 patients at 10 primary care sites, the device correctly identified the presence of more than mild diabetic retinopathy 87.4% of the time and the absence of more than mild diabetic retinopathy 89.5% of the time. Use is contraindicated in patients with a history of laser treatment, surgery, or injections in the eye and those with persistent vision loss, blurred vision, floaters, previously diagnosed macular edema, severe nonproliferative retinopathy, proliferative retinopathy, radiation retinopathy, or retinal vein occlusion. It should not be used in pregnant patients.