Joint health is notoriously difficult to track outside of clinical settings. Traditional methods for measuring joint torque—how much force a joint exerts—are often bulky, lab-bound, or too imprecise for real-world use. Now, researchers from the University of Oxford and University College London have developed a wearable device that could change that: a lightweight, AI-enabled sensor that wraps around the knee and delivers real-time data on torque, angle, and load during movement.
At the heart of the device is a piezoelectric film made from boron nitride nanotubes (BNNTs) embedded in a flexible silicone matrix. These nanotubes are prized for their mechanical strength, thermal stability, and ability to generate electrical signals when deformed. When the knee moves, the film flexes and produces a complex stream of signals that reflect the joint’s mechanical behavior. A built-in neural network processes these signals on the fly, translating them into meaningful biomechanical metrics.
What makes the system especially clever is its inverse-designed structure. The sensor has a negative Poisson’s ratio—meaning it expands laterally when stretched—allowing it to conform more naturally to the knee’s anatomy. This improves both comfort and signal fidelity, making it suitable for continuous wear during daily activities or rehabilitation exercises.
The device is low-cost, noninvasive, and designed for use in resource-limited settings. It could be a game-changer for patients with musculoskeletal conditions, athletes recovering from injury, or older adults at risk of joint degeneration. By providing real-time feedback, it enables more personalized therapy, early detection of abnormal loading patterns, and better long-term outcomes.
Press Release: AI-enabled piezoelectric wearable for joint torque monitoring: A breakthrough in joint health monitoring
Abstract from Nano-Micro Letters: AI-Enabled Piezoelectric Wearable for Joint Torque Monitoring
