EMG-Driven Musculoskeletal Modelling Framework for Virtual Simulation of Upper Limb Activation-Modulated Impairment Scenarios

Background and Objectives: Surface electromyography (EMG) is widely used to assess muscle activation. However, direct interpretation of its functional biomechanical consequences remains challenging. This study aimed to develop and evaluate an EMG-driven musculoskeletal simulation framework for investigating how controlled modifications of muscle activation patterns influence joint-level biomechanics in the upper limb. The objective was not to reproduce specific clinical pathologies but to enable systematic virtual scenario analysis of activation-dependent movement alterations. Materials and Methods: Surface EMG signals were recorded from five healthy adults (3 males, 2 females; age 22 ± 1 years) during cyclic elbow flexion/extension tasks using a wireless system (sampling frequency: 2000 Hz). Processed and normalized EMG envelopes were directly applied as prescribed neural inputs in forward dynamic simulations implemented in OpenSim, without optimization-based muscle recruitment. Controlled virtual scenarios were generated through parametric modification of activation signals to represent reduced activation capacity, increased antagonist co-activation, spasticity-like activation modulation, and tremor-like oscillatory modulation. Joint kinematics, joint moments, and movement stability were evaluated. A Movement Quality Index (MQI) was introduced as a comparative research metric integrating biomechanical performance indicators. Simulations were deterministic and analyzed descriptively. Results: Distinct activation modifications produced characteristic kinematic and kinetic responses. Reduced activation capacity decreased simulated joint moment output, increased co-activation altered joint moment timing and mechanical stability, and tremor-like oscillatory modulation generated periodic fluctuations in joint kinematics and kinetics. The MQI enabled quantitative differentiation between simulated scenarios and severity levels within the controlled modelling framework. Conclusions: The proposed EMG-driven forward dynamic simulation framework provides a methodological platform for controlled virtual scenario analysis of activation-dependent biomechanical changes. The findings highlight the sensitivity of joint-level mechanics to altered muscle activation patterns, within the deterministic modelling environment. The framework is intended for research-oriented biomechanical investigation and hypothesis testing rather than direct clinical diagnosis of neuromuscular disorders.