CRADLE: It is very difficult for doctors to do a detailed eye exam on infants. Unfortunately, babies tend to swing around, so they can't.
That's why a new one mobile application She took advantage of the photos that parents take of their children and looks for signs that they are likely to develop into a serious eye disease.
The application is the culmination of a father's attempt to find a way to find the first signs of eye disease and avoid the catastrophic loss of his child's vision.
Five years ago, NPR reported the story of Noah Shaw's son, and how he lost an eye because of him cancer.
Doctors diagnosed his retinoblastoma when he was 4 months old. To make the diagnosis, the doctors shone a light on Noah's eye, and they received an open reflection from the back of the eye. That was an indication that there were tumors there.
Noah Shaw's father is a scientist. He was wondering if he could see the same pale reflection in flashlights of his son. He had certainly seen the reflection or glow, which doctors call a "white eye," in a photo taken shortly after Noah was born.
"We had seen a white eye appear in photos at the age of 12 days," Shaw said at the time, months before his final diagnosis.
Shaw is a chemist, not an ophthalmologist, nor a computer scientist, but decided to create software that could scan photos for signs of this disease.
"If I had some software telling me 'Hi, you should check these signs at a doctor', this would have speeded up my son's diagnosis and the tumors would have been a little smaller when we arrived at them. Maybe there were fewer, ”he says.
Now, this software exists!
Together with colleagues from Baylor University in Waco, Texas, Shaw created an application called CRADLE. It uses Artificial Intelligence to find white eyes, which can be a sign of many serious eye diseases such as retinoblastoma, pediatric cataracts and Coats disease.
To test the app, they analyzed more than 50.000 photos of 40 babies. Half had no eye disease and half had been diagnosed with eye cancer or some other eye disease.
"On average, the app detected the white eye in photos taken 1,3 years before diagnosis," says Shaw.
In other words, CRADLE could give parents an early warning that something may not go well with their child. The results appear in the journal Science Advances.
The application is not perfect. Sometimes it does not show white eyes when there is a problem, and sometimes it indicates that there is a problem while it does not.
The latter is a disadvantage. Although so-called 'false positives' appear less than 1%, ophthalmologist Sean Donahue of Vanderbilt University Medical Center says it is not good enough. Donahue explains that there are about 4 million children born in the US each year. 1% false positive would mean that tens of thousands of children would go to the doctor unjustly.
Even so, Donahue is optimistic about the results of the application.
"This is exciting new technology, and I believe we will continue to detect a number of other diseases in the future," he says.
Alison Skalet, an eye specialist at Oregon Health and Science University, agrees. "There is definitely a prospect here and it makes sense to take advantage of the technology we have," he says. He expects the application to be more accurate as time goes on and its artificial intelligence to become smarter.
Bryan Shaw says in order to train the app to better recognize the white eye, it needs people who have been diagnosed with white to send it photos.
"We need more photos, especially of infants in Africa and Asia," he says. This will make the CRADLE application more relevant globally and thus hopes to save more children's eyesight.