AI-Enabled Piezoelectric Wearable for Joint Torque Monitoring

Abstract

Joint health is critical for musculoskeletal (MSK) conditions that are affecting approximately one-third of the global population. Monitoring of joint torque can offer an important pathway for the evaluation of joint health and guided intervention. However, there is no technology that can provide the precision, effectiveness, low-resource setting, and long-term wearability to simultaneously achieve both rapid and accurate joint torque measurement to enable risk assessment of joint injury and long-term monitoring of joint rehabilitation in wider environments. Herein, we propose a piezoelectric boron nitride nanotubes (BNNTs)-based, AI-enabled wearable device for regular monitoring of joint torque. We first adopted an iterative inverse design to fabricate the wearable materials with a Poisson’s ratio precisely matched to knee biomechanics. A highly sensitive piezoelectric film was constructed based on BNNTs and polydimethylsiloxane and applied to precisely capture the knee motion, while concurrently realizing self-sufficient energy harvesting. With the help of a lightweight on-device artificial neural network, the proposed wearable device was capable of accurately extracting targeted signals from the complex piezoelectric outputs and then effectively mapping these signals to their corresponding physical characteristics, including torque, angle, and loading. A real-time platform was constructed to demonstrate the capability of fine real-time torque estimation. This work offers a relatively low-cost wearable solution for effective, regular joint torque monitoring that can be made accessible to diverse populations in countries and regions with heterogeneous development levels, potentially producing wide-reaching global implications for joint health, MSK conditions, ageing, rehabilitation, personal health, and beyond.

1. Introduction

Musculoskeletal (MSK) conditions are a leading cause of disability worldwide, affecting approximately 1.71 billion people in 2019 and placing a significant burden on healthcare systems and economies [1–4]. Among the major MSK disorders, joint-related conditions such as osteoarthritis and rheumatoid arthritis significantly impact mobility and rehabilitation needs [5, 6]. These conditions not only compromise joint stability but also increase susceptibility to injuries, creating a cycle that leads to chronic pain, reduced function, and long-term disability [7]. Given the increasing prevalence of MSK disorders, particularly among high-risk populations such as the elderly and individuals with obesity, reliable methods for joint health monitoring are essential [2, 8–11].

A key factor in joint health assessment is the ability to quantify joint torque, which plays a crucial role in understanding internal joint mechanics, injury risk, and rehabilitation progress [12, 13]. Joint torque, influenced by joint angles, motion speed, external loads, and muscle activation, directly reflects internal joint stresses [13]. Excessive torque, particularly in joints like the knee, is a primary factor in injuries such as ligament tears, meniscus damage, and tendon overload [14, 15]. The knee joint, classified as a modified hinge joint, is one of the most mechanically complex load-bearing structures in the human body [16]. It primarily provides two degrees of freedom: flexion–extension and axial rotation. Flexion and extension are the dominant motions, typically ranging from ~0° to 135°, depending on individual anatomy and activity [17]. Axial rotation occurs to a limited extent when the knee is flexed, while minor lateral movements (abduction/adduction) may occur under specific conditions as passive responses to external forces. However, existing methods for assessing joint torque, including isokinetic dynamometry [18], and inverse dynamics models [19], are confined to laboratory settings, or require complex motion capture setups, limiting their feasibility for real-world applications. The direct measurement of joint torque remains a challenge, necessitating innovative wearable solutions.