Ankle-foot orthoses (AFOs) are braces worn by individuals with gait impairments to provide support about the ankle. AFOs come in a variety of designs for clinicians to choose from. However, as the effects of different design parameters on AFO properties and AFO users have not been adequately quantified, it is not clear which design choices are most likely to improve patient outcomes. Recent advances in manufacturing have further expanded the design space, adding urgency and complexity to the challenge of selecting optimal designs. A key AFO property affected by design decisions is sagittal-plane rotational stiffness. To evaluate the effectiveness of different AFO designs, we need:​ 1) a better understanding of the biomechanical effects of AFO stiffness and 2) more precise and repeatable stiffness measurement methods.

This dissertation addresses these needs by accomplishing four aims. First, we conducted a systematic literature review on the influence of AFO stiffness on gait biomechanics. We found that ankle and knee kinematics are affected by increasing stiffness, with minimal effects on hip kinematics and kinetics. However, the lack of effective stiffness measurement techniques made it difficult to determine which specific values or ranges of stiffness influence biomechanics. Therefore, in Aim 2, we developed an AFO stiffness measurement apparatus (SMApp). The SMApp is an automated device that non-destructively flexes an AFO to acquire operator- and trial-independent measurements of its torque-angle dynamics. The SMApp was designed to test a variety of AFO types and sizes across a wide range of flexion angles and speeds exceeding current alternatives.

Common models of AFO torque-angle dynamics in literature have simplified the relaxvii tionship to a linear fit whose slope represents stiffness. This linear approximation ignores damping parameters. However, as previous studies were unable to precisely control AFO flexion speed, the presence of speed effects has not been adequately investigated. Thus, in Aim 3, we used the SMApp to test whether AFOs exhibit viscoelastic behaviors over the range of speeds typically achieved during walking. This study revealed small but statistically significant effects of flexion speed on AFO stiffness for samples of both traditional AFOs and novel 3-D printed AFOs, suggesting that more complex models that include damping parameters could be more suitable for modeling AFO dynamics.

Finally, in Aim 4, we investigated the use of an active exoskeleton, that can haptically-emulate different AFOs, as a potential test bed for studying the effects of AFO parameters on human movement. Prior work has used emulation for rapid prototyping of candidate assistive devices. While emulators can mimic a physical device’s torque-angle profile, the physical and emulated devices may have other differences that influence user biomechanics. Current studies have not investigated these differences, which limits translation of findings from emulated to physical devices. To evaluate the efficacy of AFO emulation as a research tool, we conducted a single-subject pilot study with a custom-built AFO emulator device. We compared user kinematics while walking with a physical AFO against those with an emulated AFO and found they elicited similar ankle trajectories.

This dissertation resulted in the successful development and evaluation of a framework consisting of two test beds, one to assess AFO mechanical properties and another to assess the effects of these properties on the AFO user. These tools enable innovations in AFO design that can translate to measurable improvements in patient outcomes.