Feasibility Study of Real-Time Estimation of Knee Adduction Moment Using Wearable Insole Sensors and Artificial Neural Networks: A Case Report

Knee Osteoarthritis (OA) is a chronic condition affecting 645 million adults aged 40 and older worldwide. The external knee adduction moment (KAM), an indicator of medial knee load during walking, is associated with knee pain, disease severity, and structural progression. Real-time monitoring of KAM presents opportunities to retrain gait and potentially alleviate knee symptoms. However, conventional methods for calculating KAM require costly lab setups and equipment, making gait retraining by real-time KAM feedback impractical. Emerging approaches focus on using sensors, such as inertial measurement units (IMUs) and pressure sensors, to estimate KAM. This study aims to evaluate the feasibility of a low-cost method for real-time monitoring of KAM to facilitate gait retraining and reduce medial knee load, using wearable pressure-sensing insoles and artificial neural networks.