From depression and anxiety to cardiovascular disease, stress affects millions of people worldwide. Yet scientists still struggle to measure it reliably and continuously.
Today’s wearable stress monitors, such as the Fitbit Sense, often rely on electrodermal activity (EDA), which measures changes in skin conductance caused by sweat gland activity. But the technology depends on electrodes maintaining consistent contact with the skin. Conductive gels can dry out, while movement and shifting contact can interfere with readings, making long-term measurements less reliable.
A team of USC engineers introduces a new way to detect stress – using light.
Led by USC assistant professor Yasser Khan and his PhD student Md Farhad Hassan, the team developed an AI-powered smart ring that uses near-infrared light to measure subtle sweat activity beneath the skin in a study recently published.
Combining sweat and cardiovascular monitoring with embedded AI, this new ring offers a noninvasive, motion-resistant way to detect stress in real time.
Its dual-modality sensing and on-device processing also help distinguish psychological stress from movement and heat-induced sweating, outperforming conventional electrode-based wearables, according to test results from the study.
The technology is undergoing the patent application process and may become available for commercial licensing through USC’s Stevens Center for Innovation.
Started as a side project two years ago, the study resulted in a paper titled “All-optical sensing of dermal activity enables an AI-powered wearable ring for continuous stress monitoring,” and was published in Science Advances.
Khan is an Andrew and Erna Viterbi Early Career Chair and assistant professor in the USC Viterbi School of Engineering and USC Stevens School of Computing and AI’s Ming Hsieh Department of Electrical and Computer Engineering. He also has joint appointments in the Alfred E. Mann Department of Biomedical Engineering, and leads Khan Lab.
Measuring sweat with light
Why sweat gland dynamics?
When a person experiences stress, the sympathetic nervous system activates eccrine sweat glands, triggering subtle changes in sweat beneath the skin — even before visible perspiration appears, the team explained.
“Our idea is very simple. We want to look at this small sweat under the skin with light,” said Khan.
Inspired by photoplethysmography (PPG), a technique used in smartwatches to measure blood flow with light, the researchers hypothesized that specific wavelengths could also reveal changes in sweat gland activity.
Khan’s team developed a new sensing approach called “optodermal activity (ODA),” a light-based alternative to electrodermal activity and a new term coined in the study.
The system uses two near-infrared LEDs: one emitting light at 960 nanometers, which is sensitive to changes in water content associated with sweat, and another at 1,250 nanometers, which provides a stable reference. By comparing how the two wavelengths reflect beneath the skin, the ring can track changes in subsurface sweat activity without electrodes or conductive gel.
“This is the first time that someone is showing that stress can be measured with light,” Khan said.
In experiments, the optical signals showed an average 86% agreement over time with a commercial, gel-based EDA sensor during stress tests, suggesting that the new approach can capture similar physiological responses.
A smarter ring, from sensing to interpretation
Turning the sensing concept into a wearable required more than adding LEDs to a ring.
The team designed a compact, flexible electronic system that combines multiple sensors and processes their signals directly on the device.
One key innovation is the ring’s ability to capture both ODA and PPG measurements through the same optical sensing system. While ODA tracks sweat gland activity, PPG provides cardiovascular information, including heart rate and heart rate variability.
A temperature sensor and accelerometer provide additional context, helping the system distinguish psychological stress from physical movement or heat-induced sweating.
In short-term tests, the ring correctly distinguished stress from relaxation over 90% of the time, compared with less than 78% for conventional EDA. When researchers expanded the test to four conditions — stress, relaxation, physical movement and heat-induced sweating — the ring achieved 100% accuracy in the study’s test, compared with 75% for EDA.
The performance gap increased over time with extended wear.
After 12 hours, as conductive gel dried and conventional EDA measurements became less reliable, its accuracy in distinguishing stress from relaxation dropped to 57%. The ODA ring maintained approximately 80% accuracy.
The ring also incorporates Edge AI, meaning its machine-learning model runs directly on the device rather than sending raw biometric data to an external server for processing. This allows the ring to classify physiological signals locally, reducing delays while keeping sensitive personal data on the device.
The implications extend beyond personal wellness tracking.
Unlike occasional check-ins or self-reported stress levels, continuous monitoring made possible through this ring could provide insight into how stress fluctuates throughout the day, helping identify patterns that might otherwise go unnoticed.
Over time, these insights could help researchers better understand how stress affects focus, performance and well-being — and how environments can be adjusted to support healthier, more productive conditions.
Adam Frank, an assistant professor of Clinical Psychiatry and the Behavioral Sciences at Keck School of Medicine of USC, a co-author of the paper, emphasized that continuous stress monitoring with a ring can significantly improve precision psychiatry.
Stress-testing the ring
Khan plans to further validate the technology in a yearlong study involving 40 participants in an office environment, in collaboration with Burçin Becerik-Gerber, professor of Civil and Environmental Engineering, Fred Champion Chair in Civil and Environmental Engineering, and chair of the Sonny Astani Department of Civil & Environmental Engineering, and co-investigator, Gale Lucas, research associate professor at USC Viterbi and USC Stevens’ Thomas Lord Department of Computer Science and the director of the Technology Evaluation Lab at the USC Institute for Creative Technologies.
Researchers are also working to reduce interference from excessive surface sweat and expand testing to a broader range of cognitive and psychosocial stressors.
Ultimately, Khan hopes to continue developing wearable technologies that can reveal more about the relationship between the brain, the body and human behavior — with devices that are as simple and noninvasive as possible.
Journal
Science Advances
Article Title
All-optical sensing of dermal activity enables an AI-powered wearable ring for continuous stress monitoring
Article Publication Date
13-Nov-2026
COI Statement
M.F.H. and Y.K. are inventors on a US provisional patent application titled “Optodermal Wearable Device for Contactless Stress Monitoring” (serial no. 63/928,394), filed by the University of Southern California on 1 December 2025. The patent application is currently pending. The other authors declare that they have no competing interests.