A Computational Gait Model With a Below-Knee Amputation and a Semi-Active Variable-Stiffness Foot Prosthesis

Area of Science:

• Biomechanics
• Computational Modeling
• Prosthetics

Background:

• Computational musculoskeletal models are valuable for assessing biomechanical interventions like novel prostheses.
• Existing models lack limb loss-specific gait and efficient methods for variable-stiffness foot prosthesis dynamics.
• This limits the evaluation of prosthesis design parameters on gait mechanics.

Purpose of the Study:

• To develop and validate a forward simulation-capable gait model for individuals with lower-limb loss.
• To incorporate a semi-active variable-stiffness foot (VSF) prosthesis into the model.
• To enable efficient simulation of gait with advanced prosthetic devices.

Main Methods:

• A seven-segment, 28-degree-of-freedom (DoF) gait model was created.
• Forward kinematics simulations were performed using experimentally observed joint kinematics.
• The model was applied to four subjects with unilateral below-knee amputation walking with a VSF prosthesis.

Main Results:

• Model-predicted resultant ground reaction force (GRFR) showed high accuracy (R2=0.97, RMSE=7.7% BW) with optimized parameters.
• Unoptimized parameters also yielded good GRFR prediction (R2=0.93, RMSE=12% BW).
• Simulated kinematics closely matched input data (RMSE=0.23 deg, R2>0.99).

Conclusions:

• The developed methods are suitable for simulating gait in individuals with lower-limb loss.
• The model can predict GRFR for novel variable-stiffness foot prostheses.
• This facilitates user-specific optimization of prosthesis design parameters.