Wondering if your kid is dealing with an ear infection? Soon, according to researchers at the University of Washington, there’ll be an app for that. They claim to have created a simple test that uses a smartphone and folded up paper to detect one of the telltale signs of infection—fluid in the ears—with about the same or greater accuracy as a doctor.
Ear infections are one of the first health problems people tend to experience. By the age of three, most everyone has had at least one ear infection. These infections often cause fluid build-up in the ear, as can another condition called otitis media with effusion (OME). But though most infections or cases of OME go away on their own, too much or chronic fluid can cause pain or even severe complications like hearing loss.
There are already very accurate tests (around 80 to 90 percent) that can diagnose ear fluid by looking for subtle changes in the eardrum, such as tympanometry, but these need to be done at a doctor’s office and require specialized medical equipment. So the authors behind this study, published Wednesday in Science Translational Medicine, wanted to create an easier-to-use, inexpensive test that could rival that accuracy.
Their test works by having the app—called EarHealth—play chirping sounds, which are funneled through the ear canal by a folded-up piece of paper that’s taped to the phone and sits on the outer ear. These waves of sound bounce from the middle ear back to the phone, and along the way, they interact with the sounds that are still playing from the phone. The cacophony of reflected noise is then picked up by the phone’s mic, where it’s analyzed by the app. Finally, the app, based on fluctuations in the signal it receives, predicts the odds of fluid in the ear.
“It’s a little bit like tapping a wine glass,” co-lead author Justin Chan, a doctoral student at the University of Washington’s Paul G. Allen School of Computer Science & Engineering, told Gizmodo by phone. “Depending on if the glass is empty or half-full, you’ll get a different sound. So it’s the same principle here.”
Chan and his team first tested out the concept with children between the ages of 18 months and 17 years who had been admitted to Seattle Children’s Hospital, and used them to refine the app’s predictive algorithm. Half of the children had been scheduled to undergo a surgery that could treat their chronic ear fluid buildup by placing small hollow tubes inside, while the other half were getting procedures unrelated to their ears. The study’s design was meant to be a foolproof test of the app’s accuracy.
“It’s not actually easy for existing screening tools to detect middle-ear fluid. Really, the only way you can know, with complete certainty, is to undergo a surgical operation where they make an incision into the eardrum, where it can drain the fluid. So once you make that incision, you can tell for sure if there’s fluid or not,” Chan explained.
In this initial study, the app could predict whether someone had ear fluid with 85 percent accuracy, and correctly predicted if someone didn’t have fluid with 80 percent accuracy.
Because babies are most at risk for ear infections (and are less likely to be able to vocalize their discomfort if they feel sick), the refined app was then tested with 15 children between 9 and 18 months old. This time, the app correctly identified all 5 children who had ear fluid, while correctly guessing nine of the 10 children who didn’t have fluid.
Lastly, in another experiment, Chan’s team had parents try to use the app on 25 ears, following a brief tutorial. The accuracy remained as high as before, and the readings obtained by the parents closely matched the ones that doctors got when using the app. That suggests parents won’t have much trouble using the app outside of a medical setting. And most importantly, the kids across all these trials didn’t seem too flustered by having a phone funnel held up to their ear.
“The chirps are actually quite soft. And interestingly, when we played into the ears of children in the hospital, we found that they responded with smiles or laughs. It turns out the chirps have a calming effect,” said Chan.
Ideally, the app could be used as an at-home diagnostic tool by parents worried that their child’s ear tugging could be something more serious. That would let their children either avoid an unneeded trip to the doctor, or get the followup care they might need. But Chan also envisions doctors in less-developed areas using the app as a screening tool themselves, since the app is more accurate than a doctor’s simple visual inspection of the ears. Adding to the app’s potential, it was able to work on both iPhone and Android model phones, as well as with different types of paper for the funnel.
“Even in developing countries, smartphones are becoming common. So if this app can give doctors specialist-level accuracy, that could really change how ear infections are managed on a global level,” Chan said.
Chan and his co-authors have founded a company to commercialize the app, called Edus Health. He told Gizmodo that they hope to have the app cleared as a medical device by the Food and Drug Administration by the end of the year, at which point it would be widely available to the U.S. public. The company is also partnering with doctors in developing nations, with the hope that the app could soon be widely used in those countries.