UC Berkeley Engineers Develop Earbuds that Track Brain Signals Linked to Relaxation and Drowsiness

Feeling sleepy during work is a common experience, especially after lunch. However, for individuals in high-risk jobs, like driving or operating heavy machinery, drowsiness can be perilous or even fatal. Drowsy driving alone contributes to hundreds of fatal traffic accidents every year across the U.S., and the National Safety Council has highlighted drowsiness as a critical risk in industries like construction and mining.

In response to these dangers, researchers at the University of California, Berkeley, have crafted innovative earbuds that monitor the brain for early signs of drowsiness.

These specialized earbuds function similarly to an electroencephalogram (EEG), a medical test used to measure brain activity. While traditional EEG machines rely on multiple electrodes placed on the scalp, these earbuds feature internal electrodes that make contact with the ear canal to record brainwaves.

Although the electrical signals picked up by the earbuds are weaker compared to typical EEG readings, the study reveals that the "Ear EEG" platform is sensitive enough to capture alpha waves — brainwaves that become more noticeable when people close their eyes or begin to nod off.

“The inspiration struck when I bought my first Apple AirPods in 2017,” said Rikky Muller, the lead researcher and associate professor of electrical engineering and computer sciences at UC Berkeley. “I immediately thought, ‘This is a perfect platform for tracking neural signals.’ We see this as a promising technology that could not only identify drowsiness, but potentially help diagnose sleep disorders.”

Designing earbuds that simultaneously provide EEG functionality presents unique challenges. EEG electrodes must make stable contact with the skin to deliver accurate readings. In typical EEG setups, electrodes are pressed flat against the scalp, but crafting a snug, comfortable earpiece that fits various ear shapes is far more complex.

Initially, other research teams had been developing ear EEG systems using gels to improve contact or custom-molded earpieces tailored to each individual user. Muller’s team, however, aimed for a "dry" solution that could work universally — allowing anyone to use the earbuds right out of the box without extra preparation.

“I wanted to build something that a person could use daily with ease — something reusable, adaptable to various geometries, and straightforward to manufacture,” explained Ryan Kaveh, a postdoctoral scholar at UC Berkeley and co-first author of the study.

Kaveh, along with graduate student Carolyn Schwendeman, collaborated with the lab of Ana Arias at UC Berkeley to design the final version of the earpiece. Their design comes in three sizes: small, medium, and large. The earpiece is outfitted with multiple electrodes in a cantilever design to apply gentle pressure against the ear canal, ensuring good contact. Flexible electronics further enhance comfort, while a custom low-power wireless interface handles data transmission.

In a previous 2020 paper, the team demonstrated the earbuds’ ability to detect various physiological signals, including eye blinks, alpha brain waves, and the auditory steady-state response—a brain reaction to hearing a steady sound. In their latest study, they refined the earpiece design and incorporated machine learning algorithms to illustrate its potential in real-world applications.

In their tests, nine volunteers wore the earbuds while performing monotonous tasks in a dark room. Occasionally, the participants were asked to rate how drowsy they felt, and the researchers also measured their reaction times.

“Even when the signal quality wasn't perfect, we found that the earbuds could still reliably detect the onset of drowsiness with accuracy comparable to larger, more complex EEG setups,” Kaveh noted. The earbuds also proved effective at identifying drowsiness in new users, indicating they could offer 'plug-and-play' functionality for first-time wearers.

With support from the Bakar Fellowship and the Bakar Prize, Muller continues to refine the Ear EEG, looking to explore other applications beyond detecting drowsiness. These earbuds can potentially monitor heartbeats, eye movements, and jaw clenches, bringing more holistic monitoring possibilities to an everyday wearable device.

“Wireless earbuds are practically part of everyday life now,” Muller added. "That’s the beauty of the Ear EEG system. It's a natural fit for wearable tech without adding anything cumbersome to the experience.”

The study received additional support from the Ford University Research Program and a Bakar Spark Award.

Journal

Nature Communications

DOI

10.1038/s41467-024-48682-7

Method of Research

Experimental study

Subject of Research

Not applicable

Article Title

Wireless Ear EEG for Drowsiness Detection

Article Publication Date

2-Aug-2024