Injury Prediction in Competitive Runners with Machine Learning

Sofie Lovdal, Ruud den Hartigh, George Azzopardi*

*Bijbehorende auteur voor dit werk

OnderzoeksoutputAcademicpeer review

256 Downloads (Pure)

Samenvatting

Purpose: Staying injury-free is a major factor for success in sports. Although injuries are difficult to forecast, novel technologies and data science applications could provide important insights. Our purpose is to use machine learning for the prediction of injuries in runners, based on detailed training logs.
Methods: Prediction of injuries was evaluated on a new data set of 77 high-level middle and long distance runners, over a period of seven years. Two analytic approaches were applied. First, the training load from the previous seven days were expressed as a time series, with each day’s training being described by ten features. These features were a combination of objective data from a GPS watch (e.g., duration, distance), together with subjective data about the exertion and success of the training. Second, a training week was summarized by 22 aggregate features, and a time window of three weeks before the injury was considered.
Results: A predictive system based on bagged XGBoost machine learning models, resulted in Receiver Operating Characteristic curves with average Areas Under the Curves of 0.724 and 0.678 for the day and week approach, respectively. Especially the results of the day approach reflect a reasonably high probability that our system makes correct injury predictions.
Conclusions: Our machine learning-based approach predicts a sizable portion of the injuries, in particular when the model is based on training load-data in the days preceding an injury. Overall, these results demonstrate the possible merits of using machine learning to predict injuries and tailor training programs for athletes.
Originele taal-2English
Pagina's (van-tot)1522–1531
Aantal pagina's10
TijdschriftInternational journal of sports physiology and performance
Volume16
Nummer van het tijdschrift10
Vroegere onlinedatum29-apr-2021
DOI's
StatusPublished - 2021

Citeer dit