Systems for Noninvasive Assessment of Biomechanical Load in the Lower Limb
Bolus, Nicholas B.
MetadataShow full item record
Every move you make—and, yes, every step you take—is the result of action at a joint, and so proper joint function is pivotal to the way we explore and interact with the world around us. Unfortunately, joint function is often disrupted by injuries, chronic disorders, or neurological deficits, which can, in turn, disrupt quality of life. Many forms of joint dysfunction derive from adverse biomechanical loading conditions—that is, the forces and torques to which our limbs are subjected—and, thus, techniques for monitoring these loads during daily life may improve our understanding of how injuries and disorders arise and progress—and, most importantly, how best to treat them. The standard methods for assessing these loading conditions, however, are almost all benchtop-bound and confined to laboratories or clinics, so their utility in at-home or ambulatory settings—where they may be most impactful—is limited. In an attempt to address this void, in this work, we present three novel techniques for extracting information related to joint loading using a synthesis of noninvasive / wearable sensing and machine learning. First, we detail the development of an adjustable-stiffness ankle exoskeleton with multimodal sensing capabilities and use it to explore how humans interact with external elastic loading of the ankle during walking. Then, in an attempt to peer “under the skin,” we develop a novel form-factor for capturing joint sounds— the skin-surface vibrations produced by articulating structures internal to the joint—and demonstrate that these noninvasive measurements can be used to discriminate levels of axial loading at the knee. Finally, taking the concept of joint acoustics one step further, we introduce a new, active acoustics-based technique whereby the tensile loading of a particular tissue—the Achilles tendon—can be estimated by measuring the tissue’s mechanical response to a burst vibration on the skin surface. Using this approach, we are able to assess this loading state (and, by association, the net moment at the ankle) reliably across several activities of daily life, and, through a proof-of-concept study, we demonstrate how the technique can effectively translate to a fully wearable device. Collectively, the efforts reported in this thesis represent a novel, multi-path approach to assessing biomechanical loading states in the lower limb and the effects thereof. These tools and insights may serve as a basis for future development of wearable, accessible technologies for monitoring joint load during daily life, thereby reducing injury risk, tracking disease progress, assessing the efficacy of treatment, and accelerating recovery.