AI-powered smartphone scanner diagnoses diabetic retinopathy on the spot

Diabetic retinopathy, a condition that takes ophthalmologists up to a week diagnose and can lead to permanent vision loss if untreated, can now be diagnosed within minutes using a smartphone camera integrated with artificial intelligence software, according to the Ann Arbor-based University of Michigan's Michigan Medicine.

Researchers at the university's Kellogg Eye Center have devised a diabetic retinopathy scanner that combines two existing technologies: the RetinaScope handheld smartphone-turned-retinal scanner initially developed at the University of Michigan in 2016, and Woodland Hills, Calif.-based medical AI company Eyenuk's EyeArt software.

"This is the first study to combine the imaging technology with automated real-time interpretation and compare it to gold standard dilated eye examination," Yannis Paulus, MD, lead author of the study and a vitreoretinal surgeon at Kellogg, said. "And the results are very encouraging."

In a clinical study, researchers compared this automated interpretation with the findings of both human experts and traditional slit-lamp evaluation. The study showed that the AI-powered scanner detected the presence of diabetic retinopathy with nearly 87 percent sensitivity and the absence of the disease with 73 percent specificity, the latter of which was significantly higher than the human experts' specificity rates of 40 and 46 percent.

More articles about AI:
Y Combinator accelerator will prioritize biotech startups using AI for drug discovery
MIT report calls for the study of 'machine behavior' as technology advances
Robotic catheter autonomously navigates a beating heart

© Copyright ASC COMMUNICATIONS 2020. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months