The AI technology for screening diabetic retinopathy is very accurate.

A study used the images from 30,000 patient scans (120,000 images) from the English Diabetic Eye Screening Program (DESP) to identify signs of damage using the EyeArt artificial intelligence eye screening technology (Eyenuk, Inc., Los Angeles, U.S.). The results were 100% accurate for the detection of moderate to severe retinopathy that could potentially cause the loss of vision. It was 95.7% accurate while detecting issues requiring the patient to refer to a specialist’s services. The study has profoundly positive implications for diabetic retinopathy and its screening. (Medical Xpress)

By 2045, diabetic cases could number 629 million across the globe, and screening this population for eye damage would be a Herculean task. Artificial intelligence could play an immensely useful role in this endeavor and save innumerable patients their eyesight.

AI-enabled screening for diabetic retinopathy

The accuracy of AI-based screening means the technology can tremendously reduce the human effort required to screen diabetic patients every year for any signs of eye damage that could result in a loss of eyesight.

The DESP undertakes more than 2.2 million screening episodes annually. The researchers estimate that using the EyeArt machine learning technology could save £0.5 million per 100,000 screening episodes, resulting in potential annual savings of £ 10 million every year in England.

Professor Adnan Tufail, consultant ophthalmologist from Moorfields Eye Hospital and the Institute of Ophthalmology, UCL, said:

“We have shown that this validated AI software can reduce the burden of humans needing to grade diabetic eye screening images in the UK massively, by more than 5 million images per year. The technology is incredibly fast, does not miss a single case of severe diabetic retinopathy, and could contribute to healthcare system recovery post-COVID”.

“There is a very high burden on human graders required to diagnose the thousands of images every day—most of which show no signs of disease and require no further action,” said Professor Alicja Rudnicka, senior author on the research paper, from St George’s, University of London. “Our study shows that machine learning technology could safely halve the number of images that need to be assessed by humans, freeing up further funds and resources for the NHS.”

Another benefit from the technology would be the reduction in the backlog in eye screening appointments that has built up following the Covid-19 lockdown norms.

Related Story:  Amidst COVID, Eye Tests Powered by AI at Home