Estimating muscle forces and knee joint torque using surface electromyography: A musculoskeletal biomechanical model

Abstract

Surface electromyography (sEMG) is a useful tool for revealing the underlying musculoskeletal dynamic properties in the human body movement. In this paper, a musculoskeletal biomechanical model which relates the sEMG and knee joint torque is proposed. First, the dynamic model relating sEMG to skeletal muscle activation considering frequency and amplitude is built. Second, a muscle contraction model based on sliding-filament theory is developed to reflect the physiological structure and micro mechanical properties of the muscle. The muscle force and displacement vectors are determined and the transformation from muscle force to knee joint moment is realized, and finally a genetic algorithm-based calibration method for the Newton-Euler dynamics and overall musculoskeletal biomechanical model is put forward. Following the model calibration, the flexion/extension (FE) knee joint torque of eight subjects under different walking speeds was predicted. Results show that the forward biomechanical model can capture the general shape and timing of the joint torque, with normalized mean residual error (NMRE) of ∼10.01%, normalized root mean square error (NRMSE) of ∼12.39% and cross-correlation coefficient of ∼0.926. The musculoskeletal biomechanical model proposed and validated in this work could facilitate the study of neural control and how muscle forces generate and contribute to the knee joint torque during human movement.