Related Experiment Video
Updated: Jan 13, 2026
06:36
Biomechanical Analysis Methods to Assess Professional Badminton Players' Lunge Performance
Published on: June 11, 2019
11.3K
Injury risk and workload analysis in elite adolescent female volleyball players using machine learning
Théo Bouzigues1,2, Robin Candau3, Sami Äyrämö4
1Ecole Normale Supérieure de Rennes, Rennes, France - theobouzigues@orange.fr.
The Journal of Sports Medicine and Physical Fitness
|January 7, 2026
Summary
Elite female volleyball players
Area of Science:
- Sports Medicine
- Biomechanics
- Exercise Physiology
Background:
- Investigating injury predictors in elite female volleyball.
- Evaluating workload quantification, menses, and model types for injury risk.
Purpose of the Study:
- To assess if "System Training Response" (STR) workload scoring predicts injuries better than traditional methods.
- To determine if menses and external workload are key injury predictors.
- To compare linear and non-linear models for injury occurrence explanation and prediction.
Main Methods:
- Monitored 19 elite female volleyball players over a 190-day season.
- Collected internal/external workload and menses data.
- Utilized machine learning, specifically Random Forest models, for analysis.
Main Results:
- Random Forest model achieved an AUC of 0.87 for injury occurrence.
- Significant predictors included player age, menses status, and intense jump percentage.
- Cross-validation showed good generalization with an AUC of 0.74.
Conclusions:
- Intense training before performance may elevate injury risk.
- Older players may have a reduced risk of injury.
- Tailored training strategies considering physiological factors like menses are crucial for injury mitigation.
