An App That Can Catch Early Signs Of Eye Disease In A Flash

Posted by Joe Palca

Munson et al., Sci. Adv. 2019; 5 eaax 6363

An app uses a smart phone camera to detect leukocoria, a pale reflection from the back of the eye. It can be an early sign of disease. Here it appears light brown compared the healthy eye.

It’s hard for doctors to do a thorough eye exam on infants. They tend to wiggle around — the babies, that is, not the doctors.

But a new smart phone app takes advantage of parents’ fondness for snapping pictures of their children to look for signs that a child might be developing a serious eye disease.

The app is the culmination of one father’s the five-year quest to find a way to catch the earliest signs of eye disease, and prevent devastating loss of vision.

Five years ago, NPR reported the story of Bryan Shaw‘s son Noah, and how he lost an eye to cancer.

Doctors diagnosed Noah Shaw’s retinoblastoma when he was 4 months old. To make the diagnosis, the doctors shined a light into Noah’s eye, and got a pale reflection from the back of the eyeball, an indication that there were tumors there.

Noah’s father Bryan is a scientist. He wondered if he could see that same pale reflection in flash pictures his wife was always taking of his baby son. Sure enough, he saw the reflection or glow, which doctors call “white eye,” in a picture taken right after Noah was born.

“We had white eye showing up in pictures at 12 days old,” Shaw said at the time, months before his ultimate diagnosis

Shaw is a chemist, not an eye doctor nor a computer scientist, but he decided to create software that could scan photos for signs of this reflection.

“If I would have had some software in telling me ‘Hey, go get this checked out,’ that would have sped up my son’s diagnosis and the tumors would have been just a little bit smaller when we got to them. There might have been fewer,” he says.

Now, that software exists.

Along with colleagues at Baylor University in Waco, Texas, Shaw created an app called CRADLE. It uses artificial intelligence to find white eye, which can be a sign of several serious eye diseases such as retinoblastoma, pediatric cataracts, and Coats’ disease.

To test the app, they analyzed more than 50,000 pictures taken of 40 children. Half had no eye disease and half had been diagnosed with eye cancer or some other eye disease.

“On average the app detected white eye in pictures collected 1.3 years before diagnosis,” says Shaw.

In other words, the app could give an early alert to parents that something might be amiss with their child. The results appear in the journal Science Advances.

The app isn’t perfect. It sometimes misses white eye when it’s there, and sometimes says it’s there when it’s not.

That latter condition is a problem. Even though those so-called false positive occur less than 1% of the time, ophthalmologist Sean Donahue of Vanderbilt University Medical Center says that’s not good enough. Donahue explains that there are about 4 million children born in the U.S. each year. A 1% false positive rate would mean tens of thousands of children showing up at the doctor unnecessarily.

Still, Donahue is upbeat about the promise of the app.

“This is exciting new technology, and this is how I think we’re going to go for screening for a number of diseases in the future,” he says.

Alison Skalet, an eye cancer specialist at the Oregon Health and Sciences University agrees. “There’s certainly promise here and it makes sense to me to be harnessing the technology that we have,” she says. She expects the app will get more accurate as time goes on and its artificial intelligence gets smarter.

Bryan Shaw would like to see that, but to train the app to better recognize white eye, he needs people to send him pictures who’ve been diagnosed with leukocoria.

“We need more pictures. Especially from kids in Africa and Asia,” he says. That will make the app more globally relevant, and Shaw hopes, save more children’s vision.

Copyright 2019 NPR. To see more, visit https://www.npr.org.