The app is still experimental and would require clearance by the Food and Drug Administration before it could hit the market. But early data, published Wednesday in Science Translational Medicine, suggest that the smartphone can perform as well as an expensive test in a doctor’s office.
While there are many thousands of health-related apps, this one stands out because it uses the phone’s microphone and speaker to make its diagnosis.
“All you really need to do to detect ear fluid is use sound,” says Justin Chan, a graduate student at the Paul G. Allen School of Computer Science & Engineering at the University of Washington in Seattle.
To focus this sound, physicians and parents crafted a small funnel out of paper. The tip of the funnel fits into the ear canal. The app then sends short, soft pulses of sound “kind of like a bird chirping” into the ear, Chan says.
The funnel picks up the echo of that sound and the app then analyzes it. If there is fluid behind the eardrum, the echoes will sound different from those in a healthy ear. An algorithm on the phone figures it out nearly instantly.
Chan uses a wineglass as an analogy. “If a wineglass is empty or half full, tapping on it is going to produce a different sound,” he says. “And that’s exactly what we do with our tool.”
Chan is lead author of a study that included other researchers, including his close collaborator Dr. Sharat Raju, from the University of Washington and Seattle Children’s Hospital and Research Institute.
About 50 children had their ears checked with the app. Some of those children then underwent previously planned surgery on their eardrums, and that allowed doctors to verify the results of the app. The scientists report it was right about 85 percent of the time, comparable to the technology currently used in otolaryngology clinics.
Chan and his colleagues started a company to develop the app as a commercial product. He says they are in the process of seeking the FDA’s approval to market it. The agency would require more studies to gauge the app’s performance and reliability, but he is hopeful the group can gather those data by the end of the year.
“It’s very promising, but it’s too early to tell how accurate it is,” based on the newly published data, says Dr. Kenny Chan, chief of pediatric otolaryngology at Children’s Hospital Colorado. “We will have to wait and see.”
One big question is, just how useful will this be for parents and doctors?
Fluid behind the eardrum is a symptom of ear infection, but “not all fluid is an infection,” says Pamela Mudd, an ear, nose and throat specialist at Children’s National Health System in Washington, D.C. “It would be more of a test to [see if] there is something going on behind the eardrum that may be affecting my child,” rather than diagnosing an ear infection.
Doctors really need to examine a child to make that diagnosis, which is based on looking into the ear, temperature and other clinical signs, she says.
Mucus and other light fluid can accumulate behind the eardrum and not lead to infection, she says. When she examines a child’s ear and can’t tell by looking, she refers the child to a clinic where doctors use an instrument called a tympanometer, which measures fluid behind the eardrum using sound waves.
At the same time, the audiology clinic often checks for hearing loss, which helps guide treatment decisions, such as whether a child would benefit from tubes to drain built-up fluid.
Assuming the app is shown to be effective, Mudd says, she would want to talk to parents about how to interpret the results before recommending they purchase it.
“They may not have the knowledge that they need to understand what the device is telling them,” she says. The developers suggest that the app can help parents avoid a trip to the doctor’s office, but Mudd says the opposite may be the case.
“That may increase our use of the health care system” if parents take their kids to the doctor for what may be a temporary bit of fluid behind the eardrum. There may be instances where that’s appropriate, she says.
Kenny Chan, the otolaryngologist in Colorado, is also concerned about that. “To speculate that this may replace the need for a physician’s visit, I think that’s a little far-fetched,” he says.
Doctors encountered this issue after Apple marketed a watch that can identify irregular heartbeats, notes Oliver Aalami, a vascular surgeon at Stanford University who also studies mobile health applications.
“There was a lot of hype around it initially, but if you talk to the cardiologists, they were very concerned,” he says, because suddenly doctors were confronted with large numbers of worried patients, and it wasn’t clear whether all those new doctor’s appointments and interventions with drugs and tests were actually helpful.
As a result of those concerns, Apple is now conducting a big follow-up study to measure the benefits and risks of the app. Assuming the eardrum app gets FDA clearance, Aalami suspects that a similar study might be needed to figure out whether the app is on balance beneficial.
His first impression, in reading the research paper, was that the app would be more useful in a doctor’s office, both in the United States as well as in parts of the world that have less in the way of medical resources. “It may be a little too advanced for home use,” he says.
But the inventors are aiming for a home-use market. “I see it very similar to a thermometer, where if you think your child has the flu or a cold, you check their temperature several times a day,” Justin Chan says. “We think this has a similar purpose.”
He says the developers haven’t yet set a price, but they want the app to be widely available, particularly in the developing world, so it would be priced accordingly.
For this young computer scientist, this project could be a thrilling launch to his career. “I know it’s something that can touch millions of lives,” he says. “And I think that’s pretty rare in research.”