Smartphone app identifies ear infections in children
A smartphone app and a paper funnel can accurately identify pediatric patients with fluid buildup in the ear resulting from acute otitis media and otitis media with effusion, according to findings published in Science Translational Medicine.
The smartphone app uses the device’s speaker to play a series of soft, audible chirps through the funnel and into the ear canal. Depending on the way the chirps are reflected back to the phone, the app determines if fluid is present.
Justin Chan, a doctoral student at the University of Washington, told Infectious Diseases in Children that the app works similarly to tapping a wine glass.
“Depending on how much liquid is in it, you get different sounds,” he said. “In our case, we are not tapping but sending sounds and using machine learning on these sounds to detect the presence of liquid.”
An algorithm developed to detect changes in the signal and classify ears as having fluid or not was tested in children aged 18 months to 17 years (n = 98 ears). About half the children were scheduled to undergo surgery for ear tube placement (n = 48 ears), and the rest underwent surgery for different reasons and acted as controls without signs or symptoms middle ear fluid (n = 50 ears).
The app correctly identified 86% of children with serous fluid buildup, 91% of those with mucoid fluid and all children with purulent fluid in the ear.
Although some children were apprehensive about the device, the phone’s chirps “had a calming effect, causing many children to respond with smiles or laughs,” the researchers wrote.
A second cohort of children aged younger than 18 months (n = 15 ears) were enrolled to evaluate the performance of the trained algorithm. Five ears tested positive for fluid, all of which were correctly identified by the app. The app also identified nine of the 10 ears that did not have fluid.
Chan said that the technology is able to detect middle ear fluid in up to 70% wax occlusion.
Chan and colleagues wrote that parents had similar results using the app on their own children compared with physicians who used the app.
According to the researchers, the device had an overall sensitivity and specificity of 85% and 82%, respectively.
“We would like to see our technology have the widest impact possible, and we will be obtaining FDA clearance by the end of the year,” Shyam Gollakota, PhD, associate professor at the University of Washington’s Paul G. Allen School of Computer Science & Engineering, told Infectious Diseases in Children. “We have also started a health care company which aims to partner with telemedicine companies and health care organizations in developing countries to ensure our technology has global reach.” – by Katherine Bortz
Disclosures: Chan and Gollakota have equity stakes in Edus Health.