November 17, 2025
ME researchers developed a wearable device that can assess fatigue with sensors that monitor eye movements.
Researchers in ME Professor Jae-Hyun Chung's lab developed a wearable eye tracker that monitors fatigue using sensors.
“Eye movement is a window into the brain,” says ME Professor Jae-Hyun Chung, whose lab developed the device. “Our device meets the need for a lightweight, low-cost eye tracker that lasts more than 10 hours.”
Chronic tiredness — regardless of sleep — affects health, productivity and safety. It can lead to more accidents, impaired cognitive function, and a higher risk of physical and mental illness, according to the researchers.
This sensor-based eye tracker measures the percentage of eyelid closure, blink frequency, and gaze direction, which assess chronic fatigue in combination with a machine learning algorithm and a protocol to induce fatigue. It’s an improvement on a camera-based eye tracker, which is bulky, expensive and requires more computational power, or a desktop computer-based eye tracker, which doesn’t assess fatigue in a natural environment. And it could be a more accurate measure than self-reported fatigue.
The research was led by ME Ph.D. graduate Tianyi Li, in collaboration with Chung, ME researcher Changwoo Lee, ME Ph.D. student Shawn Kim, UW Medicine Associate Professor Younghoon Kwon and Professor Hojun Kim of Dongguk University in Korea.
Chung’s lab used ultrasensitive capacitive sensors composed of nanostructured electrodes that can detect surroundings through changes in the electric field it generates. The researchers rolled the carbon nanotube composites into tiny cylinders, which they then integrated into an eyeglass frame. Previous research by Chung and collaborators has found that creating sensors made of material that’s torn while wet can create a larger electric field, leading to a more effective sensor.
Using the eye tracker, participants’ eye movements were continuously monitored during daily activities to assess long-term fatigue. The fatigue levels estimated by the eye tracker showed a 73% correlation with participants’ self-reported fatigue scores.
The research team tested the device at Dongguk University’s Ilsan Hospital, where participants completed math tasks while intermittently exposed to noise to induce fatigue. In a 15-minute session, the eye-tracking system identified individuals with chronic fatigue with 74% accuracy compared with their self-reported fatigue levels, using a machine-learning model. Participants with chronic fatigue showed less responsiveness to the added cognitive challenge, demonstrating the device’s potential to detect fatigue-related performance decline.
As development progresses, the device could improve chronic fatigue evaluation and help diagnose conditions like myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). It may also be used to monitor neurological disorders, such as epilepsy. Beyond healthcare, the technology could support fatigue monitoring in emergency responders, manufacturing workers and military personnel.
“The device is the full package, with real-time data from eye tracking appearing on our computer,” says Ph.D. student Shawn Kim. “It’s cool to see how this device could potentially help people and have an impact.”