Externally Validated Machine Learning Algorithm Accurately Predicts Medial Tibial Stress Syndrome in Military Trainees; A Multi-Cohort Study.

ObjectivesMedial Tibial Stress Syndrome (MTSS) is a common musculoskeletal injury, both in sports and the military. There is no reliable treatment and reoccurrence rates are high. Prevention of MTSS is critical to reducing operational burden. Therefore, this study aimed to build a decision-making model to predict the individual risk of MTSS within officer cadets and test the external validity of the model on a separate military population. DesignProspective cohort study. MethodsThis study collected a suite of key variables previously established for predicting MTSS. Data was obtained from 107 cadets (34 females and 73 males). A follow-up survey was conducted at 3-months to determine MTSS diagnoses. Six ensemble learning algorithms were deployed and trained 5 times on random stratified samples of 75% of the dataset. The resultant algorithms were tested on the remaining 25% of the dataset and the models were compared for accuracy. The most accurate new algorithm was tested on an unrelated data sample of 123 Australian Navy recruits to establish external validity of the model. ResultsRandom Forest modelling was the most accurate in identifying a diagnosis of MTSS; (AUC = 98%). When the model was tested on an external dataset, it performed with an accuracy of 94% (F1= 0.88). ConclusionThis model is highly accurate in predicting those who will develop MTSS. The model provides important preventive capacity which should be trialled as a risk management intervention.