Evaluating the Effects of Active Ankle-Foot Orthoses on Achilles Tendon Load in Gait Rehabilitation A Comparative Musculoskeletal Study

The Achilles tendon (AT) is the strongest tendon in the human body, playing a critical role in gait by transmitting forces from the triceps surae (gastrocnemius and soleus muscles) to enable plantarflexion. Despite its strength, the AT is highly susceptible to injury due to the repetitive high-load stresses it endures during locomotor activities such as walking and running. Rehabilitation following Achilles tendon rupture or surgical repair is often prolonged and complex. Traditional protocols based on joint immobilization can lead to persistent deficits in plantarflexor strength, proprioception, and tendon stiffness, primarily due to adverse changes in muscle-tendon morphology and function—particularly affecting the soleus muscle. Active ankle-foot orthoses (AFOs) have emerged as promising tools for facilitating active rehabilitation by delivering controlled assistance that can offload the Achilles tendon and support functional recovery. By generating assistive torque in response to gait events, these devices can provide adaptive support tailored to the individual needs of the recovery process. Still, evaluating their effectiveness remains challenging due to the lack of feasible in vivo techniques for measuring AT force during dynamic movement. Invasive methods are not clinically viable, and non-invasive options like tensiometry are limited by placement complexity and postprocessing demands. Consequently, musculoskeletal modeling has become a scalable and accessible method for estimating AT load and assessing the biomechanical effects of such devices. This study aims to investigate the interaction between active AFO support and the musculoskeletal system to inform more effective rehabilitation strategies for lower limb injuries. To this end, we conducted a comparative musculoskeletal modeling analysis to estimate Achilles tendon (AT) loading during walking across five experimental conditions: walking without an AFO, with a passive AFO, and with three increasing levels of powered assistance. Surface electromyography (sEMG), kinetic, and kinematic data were collected from 14 healthy subjects walking at two speeds. Two simulation platforms, OpenSim and MyoSuite, were used to process the experimental data and estimate joint kinematics, muscle activations, and AT forces across conditions. Full time series and summary metrics were analyzed, with inverse kinematics results validated against motion capture data. Results showed strong agreement between platforms in estimating hip, knee, and ankle angles. Focus was placed on the Triceps Surae and Tibialis Anterior muscles, whose activation levels, recorded via sEMG and modeled through simulation, served as indirect indicators of AT loading. Results showed that higher powered assistance levels were associated with reduced muscle activation and tendon forces, supporting the potential of robotic AFOs to lower the mechanical demand on the Achilles tendon during gait. These findings underscore the potential of powered AFOs as tools for active rehabilitation targeting tendon recovery. Additionally, this work highlights the value of simulation-based musculoskeletal modeling using OpenSim and MyoSuite to inform both biomechanical research and clinical decision-making.