UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Technology-enabled gait monitoring and modification in real-world settings for the management of knee osteoarthritis Charlton, Jesse Marshal


For several decades biomechanical features of gait have been investigated for their relevance to musculoskeletal disease, particularly knee osteoarthritis. This is largely due to their established link between excessive or abnormally distributed ambulatory knee joint load and faster disease progression. Modifying aspects of gait kinematics, specifically the foot progression angle (FPA; a measure of in-toeing or out-toeing), can lower these joint loads and may be a tool for managing knee osteoarthritis. However, all the research both quantifying natural FPA and the work investigating how it can be modified, have relied on laboratory-based gait analysis which does not represent the typical walking environments people navigate in their daily life. In this dissertation we aimed to leverage innovative and clinically feasible technologies to move gait analysis and modification out of the laboratory and into the real-world. In the first study we demonstrated that the FPA can be measured with wearable technology both reliably and with acceptable accuracy compared to optical motion capture. We then identified the amount of gait data that needs to be collected in real-world settings to observe stable outcomes. Building from these initial studies, we performed the first characterization of the FPA in unsupervised, real-world settings over a week of community walking in a population with and without knee osteoarthritis. This showed the similarity between laboratory and real-world FPA magnitude but highlighted that laboratory measurements may underestimate the variability inherent in real-world walking. Lastly, we deployed a shoe-embedded sensor in the context of a gait modification clinical trial to monitor participants’ performance over time. This study also incorporated innovative methods, including a telerehabilitation delivery model, individualized gait modification, pre-screening for responders, and a self-directed gait modification magnitude. We found that delivering the intervention without any in-person guided practice still resulted in significant changes to the FPA, knee joint moments, and symptoms, warranting further investigation in a larger clinical trial. This dissertation demonstrated that real-world gait biomechanics can be collected with a single inertial sensor with sufficient precision to observe clinically meaningful changes, and it can be implemented in a clinical trial to monitor performance in ecologically valid environments.

Item Citations and Data


Attribution-NonCommercial-NoDerivatives 4.0 International