EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predicting noncontact injuries of professional football players using machine learning

Diogo Nuno Freitas iD, Sheikh Shanawaz Mostafa iD, Romualdo Caldeira, Francisco Santos, Eduardo Fermé, Élvio R Gouveia, Fernando Morgado-Dias

PLOS ONE · 2025

DOI: 10.1371/journal.pone.0315481 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, 2025. Verfasst von Diogo Nuno Freitas und 6 weiteren Autor:innen. Beteiligte Einrichtung: Interactive Technologies Institute (ITI/LARSyS), Funchal, Portugal. Hauptthemen laut MEDLINE: Machine Learning, Soccer, Support Vector Machine. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação para a Ciência e a Tecnologia, Agência Regional para o Desenvolvimento da Investigação, Tecnologia e Inovação, PROCiência 2020.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Noncontact injuries are prevalent among professional football players. Yet, most research on this topic is retrospective, focusing solely on statistical correlations between Global Positioning System (GPS) metrics and injury occurrence, overlooking the multifactorial nature of injuries. This study introduces an automated injury identification and prediction approach using machine learning, leveraging GPS data and player-specific parameters. A sample of 34 male professional players from a Portuguese first-division team was analyzed, combining GPS data from Catapult receivers with descriptive variables for machine learning models—Support Vector Machines (SVMs), Feedforward Neural Networks (FNNs), and Adaptive Boosting (AdaBoost)—to predict injuries. These models, particularly the SVMs with cost-sensitive learning, showed high accuracy in detecting injury events, achieving a sensitivity of 71.43%, specificity of 74.19%, and overall accuracy of 74.22%. Key predictive factors included the player’s position, session type, player load, velocity and acceleration. The developed models are notable for their balanced sensitivity and specificity, efficiency without extensive manual data collection, and capability to predict injuries for short time frames. These advancements will aid coaching staff in identifying high-risk players, optimizing team performance, and reducing rehabilitation costs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningSoccerSupport Vector MachineHumansMaleGeographic Information SystemsAdultAthletic InjuriesRetrospective StudiesYoung AdultNeural Networks, Computer

Autor:innen und Institutionen

  • Interactive Technologies Institute (ITI/LARSyS), Funchal, Portugal.
  • Faculty of Exact Sciences and Engineering, University of Madeira, Funchal, Portugal.
  • NOVA Laboratory for Computer Science and Informatics, Caparica, Portugal.
  • Department of Physical Education and Sport, University of Madeira, Funchal, Portugal.

ORCID-Profile: Diogo Nuno Freitas, Sheikh Shanawaz Mostafa

Förderung

  • Fundação para a Ciência e a Tecnologia
  • Agência Regional para o Desenvolvimento da Investigação, Tecnologia e Inovação
  • PROCiência 2020

Zitieren

APA

Freitas, D. N., Mostafa, S. S., Caldeira, R., Santos, F., Fermé, E., Gouveia, É. R., & Morgado-Dias, F. (2025). Predicting noncontact injuries of professional football players using machine learning. PLOS ONE. https://doi.org/10.1371/journal.pone.0315481

Vancouver

Freitas DN, Mostafa SS, Caldeira R, Santos F, Fermé E, Gouveia ÉR, et al. Predicting noncontact injuries of professional football players using machine learning. PLOS ONE. 2025. doi:10.1371/journal.pone.0315481

BibTeX
@article{Freitas2025,
  author = {Freitas, Diogo Nuno and Mostafa, Sheikh Shanawaz and Caldeira, Romualdo and Santos, Francisco and Fermé, Eduardo and Gouveia, Élvio R and Morgado-Dias, Fernando},
  title = {Predicting noncontact injuries of professional football players using machine learning},
  journal = {PLOS ONE},
  year = {2025},
  doi = {10.1371/journal.pone.0315481}
}
Kostenlos starten →