Abstract
Cycling, or spinning (stationary biking), is a sport that has a low incidence of acute injury and is considered a safe, low impact, cardio workout. However, the repetitive behavior of cycling makes the rider especially prone to the development of long-term injury. In literature, the cause of these injuries has been theorized to have multiple contributors, such as improper riding technique, poor compensation strategies, and overall muscle fatigue. The primary aim of this thesis is to advance the current understanding of human performance and fatigue processes by combining kinematic, kinetic, and muscle activation data from an experimental stationary bike test to identify performance metric trends, analyze EMG-based regression models of performance loss, and develop subject-specific musculoskeletal models to estimate iliotibial band syndrome (ITBS) risk factors, lateral femoral epicondyle (LFE) compression force and impingement duration. High-dimensional data was collected from sEMG-fitted volunteers as they were guided through a 1-hour stationary biking endurance routine. Motion capture and novel bike instrumentation were utilized to collect detailed kinematic and kinetic data with minimal influence on cycling performance. Following a warm-up period, pedaling resistance was manually incremented to maintain a self-reported high resistance level and reflect group stationary biking training. Volunteer fatigue levels were categorized by the extent of cadence reduction under constant cycling difficulty. Fatigued and non-fatigued performance groups exhibited significant differences, with the Fatigued group showing: greater peak hip adduction, greater lumbar flexion, greater torso and pelvic center-of-mass motion, and greater reductions in muscle activation amplitude. EMG signal shape parameters from the fatigue group were used to develop linear regression models of performance reduction. Correlation analysis and model fitting revealed a strong significance of distal motor control loss (foot plantar-/dorsi-flexion muscles), not power generation muscles, in triggering system-level performance change. OpenSim muscle path biofidelity was improved to match literature cadaveric muscle data, and estimation of ITBS risk factors was implemented. Model predictions indicate fatigued individuals have increased LFE compression duration, but decreased compression force due to kinematic changes. While osteokinematic and muscle activation compensation mechanisms were observed to maintain cadence among the non-fatigue group, increased LFE compression force was predicted with continued cycling. The investigation highlights the degradation of technique and elevated injury risk associated with fatigue and compensation processes, respectively. The identified performance-critical muscle parameters and fatigue-induced kinematic changes that can be used to inform overuse injury prevention in clinical rehabilitation, athletic training, and military applications.
