Sprint Assessment Using Machine Learning and a Wearable Accelerometer.

Field-based sprint performance assessments rely on metrics derived from a simple model of sprinting dynamics parametrized by two constants, υ0 and τ, which indicate a sprinter's maximal theoretical velocity and the time it takes to approach υ0 respectively. This study aims to automate sprint assessment by estimating υ0 and τ using machine learning and accelerometer data. To this end, photocells recorded 10 m split times of 28 subjects for three 40 m sprints while wearing an accelerometer around the waist. Features extracted from the accelerometer data were used to train a classifier to identify the sprint start and regression models to estimate the sprint model parameters. Estimates of υ0 , τ, and 30 m sprint time (t30 ) were compared between the proposed method and a photocell method using root mean square error (RMSE) and Bland-Altman analysis. The RMSE of the sprint start estimate was 0.22 s and ranged from 0.52-0.93 m/s for υ0 , 0.14-0.17 s for τ, and 0.23-0.34 s for t30 . Model-derived sprint performance metrics from most regression models were significantly (p < 0.01) correlated with 30. Comparison of the proposed method and a physics-based method suggest pursuit of a combined approach since their strengths appear to complement each other.