Principle Exploration For In-Race Metrics For Ultra-Marathon Events

Abstract

Long-distance endurance events, typically called Ultramarathons, are any distance longer than a traditional marathon of 26.2 miles. Such events cause unique stress, requiring adequate energy consumption while striving to be as lightweight and efficient as possible. Consuming enough energy during the race is necessary, but it can be difficult to determine caloric needs and timing during a long-distance run. A laboratory-based indirect calorimetry test is the gold standard for determining energy expenditure. This test has many practical limitations making it most suited for steady-state activities, and the systems are costly. Recent technological advancements include developing a wearable system by Patrick Slade, which incorporates wearable sensors on lower extremities with a microcontroller to estimate energy expenditure in real-time accurately (Slade et al., 2021). However, this system includes external wires posing a tripping hazard and possibly limiting mobility. Additionally, the system has not been tested during long-distance endurance events. Therefore, the primary aim of this study is to explore the approach and use of wireless wearable sensors on the legs to collect information during long-distance endurance events – specifically ultramarathons. The secondary aim of this study is to estimate energy expenditure during an ultra-marathon using IMUs. Participants (n = 4, age = 39.25 ± 13.45) ran a 100K ultramarathon trail race with three tri-axial inertia measurement units (IMUs) (MMS+, Mbientlab, San Francisco, CA) applied to the lateral side of the left thigh, left shank, and right shank. We collected tri-axial accelerometer and tri-axial gyroscope data throughout the race. Lower-leg kinematics were evaluated, and energy expenditure was estimated using Matlab (Mathworks, Natick, MA) (Appendix D). Lower leg kinematics and estimated energy expenditure (EEE) were estimated for four participants (mean = 173.33 ± 69.53 (W)). The calculated EEE was lower on average than expected. However, we were able to identify interesting time points in the kinematic data and EEE data that corresponded with the post-race comments from the participants. The wearable sensors used in this thesis are a low-cost wireless option, easy to apply, and used in different environments. The ability to estimate an individual’s energy expenditure accurately during long durations of physical activity outside could provide beneficial individualized biomechanical feedback to runners.