FDA Approves AI Device to Detect Diabetic Retinopathy
The software should not be used to screen for diabetic retinopathy in patients who've had laser treatment, surgery, or injection in the eye.
HealthDay News — Artificial intelligence software that can detect diabetic retinopathy has been approved by the U.S. Food and Drug Administration.
The IDx-DR program analyzes images of a patient's retinas taken with a special camera. The digital images are uploaded to a cloud server on which IDx-DR software is installed. The software provides 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." This is the first approved software that provides a screening decision without the need for a clinician to also interpret the image or results, which means it can be used by health care providers who don't typically provide eye care.
The FDA said its approval of the software was based on a clinical study of retinal images from 900 patients. IDx-DR correctly identified more than mild diabetic retinopathy 87.4 percent of the time and correctly identified patients who did not have more than mild diabetic retinopathy 89.5 percent of the time.
The software should not be used to screen for diabetic retinopathy in patients who've had laser treatment, surgery, or injection in the eye or have any of the following conditions: persistent vision loss, blurred vision, floaters, previously diagnosed macular edema, severe non-proliferative retinopathy, proliferative retinopathy, radiation retinopathy, or retinal vein occlusion, according to the FDA. The agency also said the software should not be used in pregnant women with diabetes because the software is not intended to evaluate rapidly progressive diabetic retinopathy.
The software is made by IDX LLC.