Real-Time Estimation of Lower Limb Musculoskeletal Tissue Loading Using an Electromyogram-Informed Neuromusculoskeletal Model

Abstract

Numerous musculoskeletal pathologies have been linked to altered tissue loading conditions. However, it is extremely difficult to measure in-vivo tissue loads, and internal loads are often inferred from external variables. A typical example is knee osteoarthritis (OA), with one of the main causes believed to be inappropriate loading in the tibiofemoral joint during walking. Large external adduction moments have been associated with the progression of knee OA, increased pain, and worse outcome after osteotomy surgery. However, estimates of the knee joint articular loading should also include contributions from muscle forces, which account for up to 50% of the total load. Also, muscle activation patterns differ between individuals, even when kinematics and kinetics are the same. The modification of gait based on real-time biofeedback is a non-surgical treatment that has the potential to reduce the symptoms associated with knee OA. However, current gait retraining practices focus on the reduction of the external knee adduction moment, which is not necessarily a good indicator of knee load, and it may be crucial to find better approaches to retrain gait.