Musculoskeletal model of osseointegrated transfemoral amputees in OpenSim

A Human-Prosthesis Coupled Musculoskeletal Model for Transtibial Amputees

In this paper, we present a human-prosthesis coupled full-body musculoskeletal model that integrates the dynamics of the muscle-driven human body and a motor-driven robotic prosthesis. This model can be used to perform the inverse kinematics and dynamics calculation based on measurements for amputees wearing a force-controlled or position-controlled prosthesis. As a result, we can analyze the impacts of prostheses on amputee kinetic states, such as joint torques and muscle forces. To verify the proposed model, we conducted experiments involving four transtibial amputees wearing passive prostheses and our self-developed robotic prostheses. We estimated the joint angles, joint torques, and muscle forces on the intact side and on the residual side of the subjects. The indexes reflecting the symmetry and magnitude of muscle forces were introduced to evaluate the effects of different prostheses on transtibial amputees. The indexes of muscle force magnitude indicate that the posterior thigh muscles of the residual limb exhibit significant compensation during walking. And the indexes of muscle force symmetry indicate that active prostheses with higher damping rates work better for fast walking speeds, while those with lower damping rates are more suitable for slow walking speeds. The proposed approach may offer a novel method for evaluating prostheses that considers muscle-level kinetics, thus enhancing understanding of the impact of different prostheses on the movements of amputees.

Transfemoral bone-anchored limb use changes dynamic hip muscle forces during walking

People with unilateral transfemoral amputation commonly experience pain at the residual limb-socket interface and heightened risk of musculoskeletal overuse injuries. Compensatory movement patterns acutely alleviate pain but can contribute to chronic aberrant muscle function and joint loading. Bone-anchored limbs have been shown to normalize joint loading during walking compared to socket prostheses, but their effects on muscle forces are not well understood. In this study, we compared dynamic hip muscle forces in all three planes of motion during walking before and after transfemoral bone-anchored limb implantation. Overground walking motion capture data were collected from 19 participants before (in socket prosthesis) and 12 months following bone-anchored limb implantation. Bilateral hip muscle forces were estimated during stance using subject-specific musculoskeletal models, resolved into anteroposterior, mediolateral, and superoinferior components, and compared across timepoints. After bone-anchored limb implantation, amputated-side hip abductor muscle forces were increased throughout stance (p ≤ 0.048), suggesting greater force-generating capacity of hip-spanning muscles during walking. Amputated-side hip flexor posterior muscle forces were decreased following implantation during terminal stance (p < 0.001), which may contribute to reduced anterior hip joint loading in pre-swing observed in bone-anchored limb users. Hip abductor muscle forces were more symmetric during single limb support (p < 0.034) and flexor muscle forces were more asymmetric during terminal stance (p = 0.047) following bone-anchored limb implantation. This study provides new insights of how bone-anchored limbs influence hip muscle function during walking, with implications for hip osteoarthritis development and progression in this population.

Learning to Walk With Deep Reinforcement Learning: Forward Dynamic Simulation of a Physics-Based Musculoskeletal Model of an Osseointegrated Transfemoral Amputee

This paper leverages the OpenSim physics-based simulation environment for the forward dynamic simulation of an osseointegrated transfemoral amputee musculoskeletal model, wearing a generic prosthesis. A deep reinforcement learning architecture, which combines the proximal policy optimization algorithm with imitation learning, is designed to enable the model to walk by using three different observation states. The first is a complete state that includes the agent's kinematics, ground reaction forces, and muscle data; the second is a reduced state that only includes the kinematics and ground reaction forces; the third is an augmented state that combines the kinematics and ground reaction forces with a prediction of the muscle data generated by a fully-connected feed-forward neural network. The empirical results demonstrate that the model trained with the augmented observation state can achieve walking patterns with rewards and gait symmetry ratings comparable to those of the model trained with the complete observation state, while there are no symmetric walking patterns when using the reduced observation state. This paper shows the importance of including muscle data in a deep reinforcement learning architecture for the forward dynamic simulation of musculoskeletal models of transfemoral amputees.

The gait1415+2 OpenSim musculoskeletal model of transfemoral amputees with a generic bone-anchored prosthesis

This short communication presents the gait1415+2 musculoskeletal model, that has been developed in OpenSim to describe the lower-extremity of a human subject with transfemoral amputation wearing a generic lower-limb bone-anchored prosthesis. The model has fourteen degrees of freedom, governed by fifteen musculotendon units (placed at the contralateral and residual limbs) and two generic actuators (one placed at the knee joint and one at the ankle joint of the prosthetic leg). Even though the model is a simplified abstraction, it is capable of generating a human-like walking gait and, specifically, it is capable of reproducing both the kinematics and the dynamics of a person with transfemoral amputation wearing a bone-anchored prosthesis during normal level-ground walking. The model is released as support material to this short communication with the final goal of providing the scientific community with a tool for performing forward and inverse dynamics simulations, and for developing computationally-demanding control schemes based on artificial intelligence methods for lower-limb prostheses.