aceinterpretable machine learning for injury risk prediction in university football players
Source article: SHAP-based interpretable machine learning for injury risk prediction in university football players: a multi-dimensional data analysis approach
Sports injury prediction is crucial for university football player health, yet existing research predominantly focuses on professional athletes and lacks interpretability. Using the Kaggle "University Football Injury Prediction Dataset" (800 Chinese university players), we constructed a comprehensive 18-feature evaluation system across four dimensions: basic information, training factors, physical fitness, and lifestyle habits. We systematically compared 10 machine learning algorithms. The Support Vector Machine…
Contested: both sides are scored from claims and sources, not community votes.
"Army Football Training Camp" by West Point - The U.S. Military Academy is licensed under CC BY 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/2.0/.
Researchers built an 18-feature model across basic information, training, fitness and lifestyle dimensions for 800 Chinese university football players from a Kaggle dataset, comparing 10 algorithms and finding SVM best at 95.6% accuracy with SHAP highlighting stress, sleep and balance as key predictors.
The result suggests lifestyle and psychological factors may be more predictive than fitness alone for this student-athlete population, but because the work relies on one retrospective dataset without external testing, its real-world prevention value remains unproven pending prospective validation.
- Compared 10 machine learning algorithms on Kaggle University Football Injury Prediction Dataset of 800 Chinese university players
- Built 18-feature system across basic information, training factors, physical fitness, and lifestyle habits
- SHAP analysis ranked stress level 0.10, sleep duration 0.09, and balance ability 0.08 as most important, with lifestyle factors outweighing physical fitness
SVM-based model using 18 features across four dimensions predicted injury risk in 800 university football players with 95.6% accuracy and 99.2% ROC-AUC, with SHAP identifying stress, sleep and balance as top factors.
Model was developed and tested on a single Kaggle dataset of 800 Chinese university players without external validation, limiting generalizability and preventing immediate clinical deployment.
The rundown
The study used the Kaggle University Football Injury Prediction Dataset covering 800 Chinese university players and evaluated 10 algorithms, with SVM outperforming others on accuracy, F1-score and ROC-AUC.
Interpretability analysis found psychological stress positively correlated with injury risk while adequate sleep and balance showed protective effects, and noted lifestyle factors outweighed traditional physical fitness indicators.
Single-dataset design without external validation limits generalizability and requires prospective validation before clinical deployment.
Sources
- Peer-reviewedScientific Reports2025-11-17
How should this claim be treated?
The debate