Your earbuds might soon be able to tell you how disgusting your ears are
- Justin Caffier
A new study presented in Portland explains how AI might help improve our ear health.
For many of us, our earbuds are nothing short of an extension of our body. With the practice of wearing a Bluetooth earpiece in public now no longer associated with just self-important business types or Boomers lacking volume control, the former stigmas have been eradicated. Today, it’s nothing to go through our workdays, workouts, commutes, and chores at home with the little plastic barnacles stuck in our ear canals, taking them out only to recharge the battery or show our good manners.
Whether or not we’re ready to admit it, earbuds are currently serving as the most common bridge between the flesh and blood humans of the past and whatever hyper-augmented cyborg future awaits us. Though once merely a vehicle to pump in our terrible music tastes, earbuds have evolved to provide a host of other useful functions over the decades. Today’s top of the line models connect phone calls and FaceTimes, read incoming texts to us, and block out all external noise whenever desired. (And yes, they of course still play our awful favorite songs. But as it is 2022, we’ve added a few terrible podcasts to the rotation as well.) And according to a recent study presented in Portland, Oregon at the Association for Computing Machinery International Conference on Mobile Systems, Applications, and Services (MobiSys), the digital Swiss Army Knives jammed in our ears might soon get an additional tool: ear hygiene monitoring.
The published research, conducted by a team at the University at Buffalo, lays out a strong case for the inclusion of what they’re calling “EarHealth” in the devices’ current bag of tricks. By using Bluetooth earbuds connected to a smartphone with deep learning capabilities, the team was able to develop a system that accurately detects three of the most common ear problems plaguing the world today: earwax blockage, ruptured eardrums, and middle ear infections called “otitis media.” And since the proliferation of earbuds and wearing them all day has directly contributed to an increase of these maladies, it only makes sense that they should start accounting for their crimes.
EarHealth works like a bat’s echolocation. It sends a chirp out into the wearer’s ear canal, recording how the reverberations bounce around and then using the data to map out their unique shape once enough data has been recorded. Once enough chirps have been sent out to determine the canal’s true baseline shape, subsequent chirps are easily able to spot worrisome changes to the landscape and alert the wearer accordingly.
Using a sample group of 27 healthy patients, 22 with ruptured drums, 18 with wax blockages, and 25 with otitis media, researchers were able to train EarHealth to spot these conditions with 82.6% accuracy.
In a statement published by the University at Buffalo’s news center, the study’s lead author, Zhanpeng Jin, Ph.D., says that because “people worldwide are living longer, and [because of] the prevalence of headphones, it is more important than ever to monitor one’s ear health.” After touting the system’s affordability, effectiveness, and ease of use for users, he added that because EarHealth “has the potential to detect these conditions very early, it could greatly improve health outcomes for many people.”
The EarHealth team is currently in their refining era, tightening up those accuracy percentages and planning for the inclusion of future variables like the detection of ear hair and eardrum inflammation. But even in its current state, it seems like an obvious buy for a tech company that’s already pushing a suite of products that synergistically monitor one’s health. So when those 4th generation AirPod ads start dropping and subtly shaming you for not already keeping tabs on your aural hygiene, try to keep in mind that the EarHealth crew started that endeavor with noble intention.