UC Berkeley researchers have developed earpieces that detect brain activity associated with relaxation and drowsiness.

Ryan Kaveh/UC Berkeley

It’s important to remember the following text: “Many people experience sleepiness at work, particularly after a large meal. However, drowsiness can pose a serious danger for individuals in roles that involve driving or operating heavy machinery, potentially leading to fatal accidents. In the U.S., drowsy driving results in hundreds of deadly vehicle crashes each year, and the National Safety Council has identified drowsiness as a significant risk in the construction and mining industries.”

Engineers at the University of California, Berkeley, have developed prototype earbuds capable of detecting signs of drowsiness in the brain to protect drivers and machine operators from the dangers of falling asleep at the wheel.

The earbuds detect brain waves similar to an electroencephalogram (EEG), a test doctors use to measure electrical activity in the brain. While most EEGs use electrodes attached to the head to detect brain waves, the earbuds have built-in electrodes designed to make contact with the ear canal for the same purpose.

The electrical signals detected by the earbuds are smaller than those picked up by a traditional EEG. However, a new study by the researchers shows that their Ear EEG platform is sensitive enough to detect alpha waves, a pattern of brain activity that increases when you close your eyes or start to fall asleep.

“I was inspired when I bought my first pair of Apple’s AirPods in 2017. I immediately thought, ‘What an amazing platform for neural recording,'” said study senior author Rikky Muller, an associate professor of electrical engineering and computer sciences at UC Berkeley. “We believe that this technology has many potential uses. Classifying drowsiness is a good indicator that the technology can be used to classify sleep and even diagnose sleep disorders.”

Using an earbud as an EEG electrode presents several practical challenges. The electrodes must maintain good contact with the skin to obtain an accurate EEG reading. This is relatively simple to achieve with traditional EEGs, which use flat metal electrodes attached to the scalp. However, it is more challenging to design an earbud that will fit securely and comfortably in ears of various sizes and shapes.

“When Muller’s team began working on the project, other groups developing Ear EEG platforms used wet electrode gels to ensure a good seal between the earbud and the ear canal or create custom-moulded earpieces for each user. She and her team aimed to design a dry and universally applicable model, allowing anyone to insert them in their ears and obtain reliable readings.”

“My personal goal was to create a device that could be used daily by individuals who would greatly benefit from it,” explained Ryan Kaveh, a postdoctoral scholar at UC Berkeley and co-first author of the study. “To achieve this, I understood that it needed to be reusable, adaptable to various people, and simple to manufacture.”

Kaveh co-led the study with graduate student Carolyn Schwendeman and collaborated with Ana Arias’s lab at UC Berkeley to design the final earpiece in three sizes: small, medium, and large. The earpiece incorporates multiple electrodes in a cantilevered design that applies gentle outward pressure to the ear canal and uses flexible electronics to ensure a comfortable fit. The signals are read through a custom, low-power, wireless electronic interface.

In a 2020 paper, researchers demonstrated that these earpieces can detect various physiological signals, such as eye blinks, alpha brain waves, and the auditory steady-state response, which is the brain’s reaction to hearing a constant pitch. In the new study, they enhanced the earpiece design and utilized machine learning to show how the earpieces could be applied in real-world scenarios.

During the experiment, nine volunteers were asked to wear earpieces while performing mundane tasks in a dimly lit room. Periodically, the volunteers rated their level of drowsiness and measured their response times.

“We discovered that even when the signal quality from the earpieces appeared to be worse, we were still able to accurately detect the onset of drowsiness, just as effectively as much more complicated, bulky systems,” Kaveh explained. The earpieces also maintain their accuracy when identifying drowsiness in new users, a feature that makes them suitable for use right out of the box.”

“Developed with the support of the Bakar Fellowship and the Bakar Prize, Muller is continuing to refine the design of the Ear EEG and explore other potential applications of the device. In addition to recording EEG signals, the device can also record other signals such as heartbeats, eye movements, and jaw clenches.”

“We constantly wear wireless earbuds,” Muller explained. “That’s what makes Ear EEG so compelling. It doesn’t require anything extra.”