An evaluated gait-driven finite element framework for transfemoral limb–socket interface biomechanics and socket design evaluation

Abstract

Abstract Background Residual limb–socket interface mechanics are central to transfemoral prosthetic comfort and stability, but their dependence on socket material, regional pressure loading, interfacial slip ( δ slip ), and interfacial friction ( µ ) remains insufficiently quantified under physiological gait loading. This study aimed to develop a gait-driven FE framework for controlled pre-fabrication comparison of material- and interface-dependent socket responses under stance-phase loading. Methods A unilateral transfemoral limb–socket FE model was developed using trial-specific stance-phase gait data digitised from the dynamic residual limb–socket coupling experiments reported by Tang et al. in 2015. Vertical ground reaction forces ( v GRFs) and residual femur–socket rotations were applied as model inputs, while axial coupling length was used for model evaluation. Four trial-specific simulations were compared with experimental measurements using Pearson correlation coefficient ( r P ), RMSE, and NRMSE. The model was then applied to a controlled comparison of ten candidate socket materials under identical geometry, loading, and contact conditions, quantifying structural safety ( R ), contact pressure, δ slip , and shear stress ( τ max ). µ was further varied to quantify its influence on residual-limb soft-tissue stress and coupled contact behaviour. Results The model reproduced the main temporal pattern of stance-phase axial coupling in the analysed single-subject dataset, showing very strong temporal association and acceptable point-wise discrepancy ( r P = 0.906; 95% CI 0.848–0.943; RMSE 3.32 mm; NRMSE 12.23%; p < 0.001). Higher-stiffness and higher-strength sockets provided greater structural safety margins. Contact pressure exhibited pronounced spatiotemporal non-uniformity, with posterior (P) and adjacent regions carrying the highest loads. Higher socket stiffness delayed high-slip onset by 22.59% of the gait cycle and reduced slip build-up rate by approximately 57.01%, whereas τ max showed only negligible dependence on stiffness (< 2%). Soft-tissue maximum von Mises stress ( σ vM, max ) exhibited a non-monotonic dependence on µ , reaching a model-specific near-minimum across µ = 0.6–0.7. Conclusion Within this model-evaluation scope, the gait-driven FE framework provides a quantitative pre-fabrication basis for comparing material- and interface-dependent socket responses. Broader patient-specific interpretation requires further validation using additional subjects and direct interface measurements.

The authors' abstract, as published at the source. Journal of NeuroEngineering and Rehabilitation, 2026 · DOI ↗