Improving Human Balance with Wearable Devices

Abstract

This study investigates the use of an optical-based balance sensor and a musclemimetic wearable robot to evaluate and improve balance in elderly people with various health conditions. We analyzed data from 149 subjects to extract critical temporal and frequential features related to balance, such as center of pressure (CoP), center of gravity (CoG), and theta angles, to categorize them into five distinct levels (i.e., levels 0,1,2,3, and 4). A correlation analysis between balance sensor features and gastrocnemius lateralis (GL) muscle maximum voluntary contraction (MVC) confirmed our hypothesis that subjects with stronger GL muscles maintained better balance. Hence, we developed a wearable muscle-mimetic robot to compensate for the weakened GL muscle. The GL muscle-mimetic ankle robot, which mimics muscle mechanical properties and uses human intrinsic physiological signals for control, simultaneously relaxes and contracts with the GL muscle to enhance postural stability and balance. Furthermore, testing the wearable robot with the balance sensor demonstrated promising results in enhancing balance, as evidenced by the decreased variance of CoP when the robot was worn. Our findings suggest that the optical-based balance sensor and muscle-mimetic wearable robot provide an effective approach for assessing and improving balance in elderly individuals, with potential implications for reducing fall risk and improving postural stability in aging populations.