Hans Van Eetvelde1, Luciana D Mendonça2,3,4, Christophe Ley5, Romain Seil6, Thomas Tischer7. 1. Department of Applied Mathematics, Computer Science and Statistics, Ghent University, Krijgslaan 281-S9, 9000, Ghent, Belgium. Hans.VanEetvelde@UGent.be. 2. Graduate Program in Rehabilitation and Functional Performance, Universidade Federal Dos Vales Do Jequitinhonha E Mucuri (UFVJM), Diamantina, Minas Gerais, Brazil. 3. Department of Physical Therapy and Motor Rehabilitation, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium. 4. Ministry of Education of Brazil, CAPES Foundation, Brasília, Distrito Federal, Brazil. 5. Department of Applied Mathematics, Computer Science and Statistics, Ghent University, Krijgslaan 281-S9, 9000, Ghent, Belgium. 6. Department of Orthopaedic Surgery, Centre Hospitalier Luxembourg and Luxembourg Institute of Health, Luxembourg, Luxembourg. 7. Department of Orthopaedic Surgery, University of Rostock, Rostock, Germany.
Abstract
PURPOSE: Injuries are common in sports and can have significant physical, psychological and financial consequences. Machine learning (ML) methods could be used to improve injury prediction and allow proper approaches to injury prevention. The aim of our study was therefore to perform a systematic review of ML methods in sport injury prediction and prevention. METHODS: A search of the PubMed database was performed on March 24th 2020. Eligible articles included original studies investigating the role of ML for sport injury prediction and prevention. Two independent reviewers screened articles, assessed eligibility, risk of bias and extracted data. Methodological quality and risk of bias were determined by the Newcastle-Ottawa Scale. Study quality was evaluated using the GRADE working group methodology. RESULTS: Eleven out of 249 studies met inclusion/exclusion criteria. Different ML methods were used (tree-based ensemble methods (n = 9), Support Vector Machines (n = 4), Artificial Neural Networks (n = 2)). The classification methods were facilitated by preprocessing steps (n = 5) and optimized using over- and undersampling methods (n = 6), hyperparameter tuning (n = 4), feature selection (n = 3) and dimensionality reduction (n = 1). Injury predictive performance ranged from poor (Accuracy = 52%, AUC = 0.52) to strong (AUC = 0.87, f1-score = 85%). CONCLUSIONS: Current ML methods can be used to identify athletes at high injury risk and be helpful to detect the most important injury risk factors. Methodological quality of the analyses was sufficient in general, but could be further improved. More effort should be put in the interpretation of the ML models.
PURPOSE: Injuries are common in sports and can have significant physical, psychological and financial consequences. Machine learning (ML) methods could be used to improve injury prediction and allow proper approaches to injury prevention. The aim of our study was therefore to perform a systematic review of ML methods in sport injury prediction and prevention. METHODS: A search of the PubMed database was performed on March 24th 2020. Eligible articles included original studies investigating the role of ML for sport injury prediction and prevention. Two independent reviewers screened articles, assessed eligibility, risk of bias and extracted data. Methodological quality and risk of bias were determined by the Newcastle-Ottawa Scale. Study quality was evaluated using the GRADE working group methodology. RESULTS: Eleven out of 249 studies met inclusion/exclusion criteria. Different ML methods were used (tree-based ensemble methods (n = 9), Support Vector Machines (n = 4), Artificial Neural Networks (n = 2)). The classification methods were facilitated by preprocessing steps (n = 5) and optimized using over- and undersampling methods (n = 6), hyperparameter tuning (n = 4), feature selection (n = 3) and dimensionality reduction (n = 1). Injury predictive performance ranged from poor (Accuracy = 52%, AUC = 0.52) to strong (AUC = 0.87, f1-score = 85%). CONCLUSIONS: Current ML methods can be used to identify athletes at high injury risk and be helpful to detect the most important injury risk factors. Methodological quality of the analyses was sufficient in general, but could be further improved. More effort should be put in the interpretation of the ML models.
Entities:
Keywords:
Injury prediction; Injury prevention; Machine Learning; Sport injury
Authors: Francisco Ayala; Alejandro López-Valenciano; Jose Antonio Gámez Martín; Mark De Ste Croix; Francisco J Vera-Garcia; Maria Del Pilar García-Vaquero; Iñaki Ruiz-Pérez; Gregory D Myer Journal: Int J Sports Med Date: 2019-03-14 Impact factor: 3.118
Authors: David L Carey; Kay M Crossley; Rod Whiteley; Andrea Mosler; Kok-Leong Ong; Justin Crow; Meg E Morris Journal: Med Sci Sports Exerc Date: 2018-11 Impact factor: 5.411
Authors: N F N Bittencourt; W H Meeuwisse; L D Mendonça; A Nettel-Aguirre; J M Ocarino; S T Fonseca Journal: Br J Sports Med Date: 2016-07-21 Impact factor: 13.800
Authors: João Gustavo Claudino; Daniel de Oliveira Capanema; Thiago Vieira de Souza; Julio Cerca Serrão; Adriano C Machado Pereira; George P Nassis Journal: Sports Med Open Date: 2019-07-03
Authors: A L Rahlf; T Hoenig; J Stürznickel; K Cremans; D Fohrmann; A Sanchez-Alvarado; T Rolvien; K Hollander Journal: BMC Sports Sci Med Rehabil Date: 2022-04-26