Multiple Sclerosis (MS) is a progressive neurodegenerative disease characterizedby demyelination and axonal damage throughout the central nervous system. The resulting sensory, motor, and cognitive impairments lead to significant mobility challenges, frequently resulting in falls. In addition to potential injury, falls have deleterious effects on independence and quality of life. Currently, fall risk is assessed through episodic clinical visits that fail to capture symptom fluctuations, and interventions are prescribed reactively after falls occur. Wearable sensors combined with mobile health platforms offer opportunities for continuous monitoring and real-time intervention, but key technological and scientific gaps limit their implementation. The purpose of this work was to identify neuromuscular biomarkers of fall risk, develop methods for long-duration monitoring, and evaluate real-time interventions to modify fall-related biomechanics in Persons with Multiple Sclerosis (PwMS). The first contribution of this work was the identification of tibialis anterior (TA)muscle activity patterns as neuromuscular biomarkers of fall risk. Using wearable surface electromyography (sEMG) during gait, specific TA parameters were found to correlate with clinical measures of disability and successfully predict fall status through logistic regression models. This establishes a direct link between muscle activity patterns and fall risk that reveals more specific fall-risk parameters. The second contribution was an adaptive normalization algorithm that enablessEMG normalization in free-living environments. Traditional normalization methods resulted in widespread erroneous data during long-duration recordings, while the adaptive approach reduced erroneous data by updating reference values based on changing environmental conditions. This algorithm addresses a fundamental barrier to translating sEMG monitoring from laboratory to daily life. The third contribution demonstrated that PwMS can produce immediate, targetedbiomechanical changes in response to visual feedback delivered via a mobile application. Point-of-choice prompts successfully modified stance width during walking and trunk flexion during sit-to-stand transitions—both key factors in maintaining dynamic stability. High user acceptance rates suggested strong potential for clinical translation. Overall, this body of work establishes the scientific and technological foundationfor continuous, proactive fall prevention in PwMS. By bridging the gap between neuromuscular monitoring and real-time intervention, this dissertation enables a paradigm shift from reactive clinical care to personalized, just-in-time support in daily life where falls actually occur.
Samantha R. Fox (Wed,) studied this question.