L. Zumeta Olaskoaga, D. Lee, J. Larruskain, E. Bikandi, I. Setuain, J. Lekue
Injuries are a common occurrence in professional sports and modelling and understanding injury patterns are of increasing interest in order to enable individually adequate training control for athletes. In this work, 22 female professional football players were followed-up for a 2017-2018 season. Individual player exposure (training and competition minutes) and time-loss injuries were recorded by the club’s medical staff. Players completed biomechanical and functional conditioning screenings to identify movement asymmetries that may predispose players to injury. The players completed these screenings at pre-season and mid-season. In the context of recurrent time-to-event data, i) we compare several variable selection techniques (including survival tree-based and penalized cox regression methods), ii) discuss the use of frailty Cox models, to account for unobserved heterogeneity and within-player dependence and iii) perform a simulation study to evaluate the robustness of the models.
Palabras clave: Shared frailty models, Regularized Cox methods, Sports injury prevention
Programado
Sesión Invitada Análisis de datos en el deporte
8 de junio de 2022 17:20
Sala Audiovisuales
