Analysis and assessment of injury risk in female gymnastics: Bayesian Network approach

This paper presents a Bayesian network (BN) model for estimating injury risk in female artistic gymnastics. The model illustrates the connections betweenunderlying injury risk factorsthrough a series ofcausal dependencies. The quantitativepart of the model ? the conditional probability tables, are determined using ТNormal distribution with parameters, derived by experts. The injury rates calculated by the network are in an agreement with injury statistic data and correctly reports the impact of various risk factors on injury rates. The model is designed to assist coaches and supporting teams in planning the training activity so that injuries are minimized. This study provides important background for further data collection and research necessary to improve the precision of the quantitative predictions of the model.