Related Experiment Videos
Injury prediction in elite women's football: an integrative machine learning-based decision-support framework
Manuel Huth1,2, Berta Canal-Simón3,4, Eva Ferrer5,6
1Life and Medical Sciences (LIMES) Institute, University of Bonn, Bonn, Germany.
NPJ Digital Medicine
|July 8, 2026
Summary
This study introduces a new injury prediction framework for elite sports, improving player availability by accounting for cumulative risk and injury severity. The novel approach enhances decision-making for injury prevention and optimizes long-term player performance.
Area of Science:
- Sports Medicine
- Data Science in Sports
- Injury Epidemiology
Background:
- Elite sports injuries negatively impact performance, careers, and finances.
- Current injury prediction methods lack accuracy due to unaddressed cumulative risk, severity, calibration, and decision thresholds.
Purpose of the Study:
- To develop and validate a novel injury prediction framework for elite athletes.
- To improve injury prevention strategies and player availability in professional sports.
Main Methods:
- Integrated survival analysis for risk accumulation with machine learning.
- Incorporated probability beta calibration and statistical decision theory.
- Utilized a unique four-season dataset from FC Barcelona's women's team.
Main Results:
- The novel framework demonstrated superior discrimination ability compared to standard classifiers.
- Identified fatigue-related measures as significant predictors of injury.
- Showcased improved player availability through flexible decision thresholds.
Conclusions:
- The developed framework offers a significant advancement in predicting elite sports injuries.
- The approach is scalable and transferable to various sports, bridging research and practical application.
- Empowers sports organizations to optimize player performance and long-term outcomes.