Prediction models for musculoskeletal injuries in professional sporting activities: A systematic review

Abstract

The purpose of this systematic review was twofold: (a) identify prediction models for musculoskeletal injuries during the participation of professional sporting activities and (b) evaluate these models by their predictive performances. A systematic review of the PubMed and Embase databases was performed using specific search terms selected according to the PRISMA guidelines. Ten studies met the eligibility criteria and were included. The most commonly employed data component for data pre‐processing was body composition data (most commonly, body mass and height), followed by player profile (most commonly, age followed by position). The most common machine learning technique for data processing was the decision tree, followed by logistic regression. The median AUC of the best performing models indicated per study was 0.75 (0.16), median sensitivity/recall was 0.78 (0.15), median specificity was 0.81 (0.27), and median precision was 0.53 (0.13). The performance of prediction models in the literature has been poor, caused by a fundamental difficulty in discovering real effects in small sample sizes with low injury rates. A better understanding of how training and game exposure is associated with the data components for data pre‐processing and ultimately associated with injury is vital for the future development of robust injury prediction models.

Wearable Technology and Analytics as a Complementary Toolkit to Optimize Workload and to Reduce Injury Burden

“…To take a step further, multiple studies have investigated ML techniques to detect professional athletes at high-risk of injury using workload as the main injury determinant. These prediction models, motivated by the data suggesting a workload-injury risk association, apply data processing techniques and ML to estimate injury probability (Seow et al, 2020 ). Colby et al, investigated lower body non-contact injuries in professional Australian football athletes and measured sRPE and GPS-derived total distance, sprinting distance, and maximal velocity over three seasons and found that minimal athlete exposures to high velocity efforts (85% maximal velocity) over the previous 8 weeks were associated with significantly greater injury risk than athletes with a greater number of high velocity efforts (Colby et al, 2014 ).…”

“…Predictive models have the potential to act as an automated data analyst capable of providing insight into the athlete's condition. The performance of predictive models in the literature thus far has been poor due, in part, to their small sample sizes with low injury rates, but also because our incomplete understanding of the determinants of injury and how these variables behave in a dynamic system of athlete performance (Seow et al, 2020 ).…”

Improving prediction model systematic review methodology: Letter to the Editor

“…Dear Editor, In their recently published paper, Seow et al 1 carried out a systematic review of musculoskeletal injury prediction models in professional sport and military special forces. Their review encompassed a comprehensive search that included both conference and published papers, used a standardized musculoskeletal injury definition that was informed by the literature, and included both statistical and machine learningbased models.…”

“…Hence, the NOS is a blunt instrument to assess risk of bias in these studies. The tool that should have been used to assess the risk of bias in the review by Seow et al 1 is the Prediction model Risk Of Bias Assessment Tool (PROBAST), 2 which includes 20 signaling questions over four domains (participants, predictors, outcome, and analysis), to cover key aspects of prediction model studies. Furthermore, when designing a systematic review of prediction model studies, the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) checklist 3 provides detailed guidance to help authors in developing their systematic review questions relating to prediction models, extracting pertinent prediction model data, and appraising prediction model studies.…”

A Narrative Review for a Machine Learning Application in Sports: An Example Based on Injury Forecasting in Soccer

“…As a matter of fact, this simplification hides the complexity of the training stimuli, not allowing for the detection of complex patterns in training workloads linked to injuries [ 16 , 17 , 18 ]. For this reason, the literature concerning multidimensional models focused on predicting injury is growing fast [ 19 ]. In this review, the authors state that the performance of prediction models was still poor due to the difficulty to solve this very complex modeling problem.…”