نشریه علمی مهندسی پزشکی زیستی

Integration of OpenSim Musculoskeletal Modeling with Surface EMG for Real-Time Knee Torque Estimation in Rehabilitation Applications

Document Type : Full Research Paper

Authors

1 University of Tabriz

2 Department of Mechatronics Engineering, University of Tabriz

10.22041/ijbme.2026.2093504.2044
Abstract
Accurate estimation of joint torque is essential for the control and evaluation of active rehabilitation systems. While surface electromyography (sEMG) offers a non-invasive measure of muscle activation, musculoskeletal modeling enables the estimation of internal joint moments without reliance on costly dynamometers. In this study, we present an integrated framework that combines real-time 2-channel sEMG recordings and knee angle measurements with a subject-specific OpenSim musculoskeletal model of the seated leg, interfaced through MATLAB. Experimental data from a healthy participant were used to drive the model, estimate individual muscle forces, and compute the net knee extension torque. To this end, a forward dynamic simulation framework, where the muscle-driven excitations were derived from the measured sEMG signals to actuate the knee extensors and flexors in the OpenSim model. The estimated peak torque (~25 N·m), angular impulse (~50 N·m·s), and average torque (~12.5 N·m) were found to be consistent with physiological ranges reported in the literature. The proposed framework provides a practical, cost-effective, and scalable solution for real-time knee torque monitoring, with significant potential for integration into sensor-driven rehabilitation devices and biomechanical assessment protocols.

Keywords

Subjects


Articles in Press, Accepted Manuscript
Available Online from 12 August 2026

  • Receive Date 05 July 2026
  • Revise Date 08 August 2026
  • Accept Date 12 August 2026