3D Fiber-Reinforced Forward Musculoskeletal Modeling of the Lower Limb Using HD-sEMG and CFD-FE Coupling.

Abstract

Accurate representation of three-dimensional muscle mechanics is critical for constructing physics-informed digital twins of the human musculoskeletal system. The purpose of this study was to develop a forward musculoskeletal modeling framework that directly links electrophysiological activation to three-dimensional muscle deformation and force transmission. Subject-specific muscle geometry and fiber architecture were reconstructed in three dimensions. Spatially resolved activation patterns extracted from HD-sEMG were mapped onto the muscle domain to drive active contraction. An anisotropic hyperelastic constitutive model with embedded fiber reinforcement was implemented through a user-defined material subroutine in Abaqus to represent both passive and active muscle behavior. Forward simulations of hip abduction, flexion, and extension were conducted. The proposed framework reproduced physiologically realistic muscle deformation and non-uniform fiber-level stress-strain distributions during active contraction. Muscle activation drove coordinated joint motion in a forward modeling manner without requiring inverse dynamics-based optimization. Predicted hip joint kinematics showed reasonable agreement with motion capture-derived joint angles across all simulated movements, while the simulated muscle contraction velocities remained within physiological ranges reported in the literature. By directly coupling HD-sEMG-derived activation with three-dimensional fiber-reinforced muscle mechanics, this framework overcomes key limitations of conventional musculoskeletal models. The proposed approach enables subject-specific, predictive simulation of muscle function and provides a robust foundation for applications in surgical planning, rehabilitation assessment, and implant-tissue interaction analysis.