Comparison of optimized machine learning approach to the understanding of medial tibial stress syndrome in male military personnel.

PURPOSE: This study investigates the applicability of optimized machine learning (ML) approach for the prediction of Medial tibial stress syndrome (MTSS) using anatomic and anthropometric predictors. METHOD: To this end, 180 recruits were enrolled in a cross-sectional study of 30 MTSS (30.36 ± 4.80 years) and 150 normal (29.70 ± 3.81 years). Twenty-five predictors/features, including demographic, anatomic, and anthropometric variables, were selected as risk factors. Bayesian optimization method was used to evaluate the most applicable machine learning algorithm with tuned hyperparameters on the training data. Three experiments were performed to handle the imbalances in the data set. The validation criteria were accuracy, sensitivity, and specificity. RESULTS: The highest performance (even 100%) was observed for the Ensemble and SVM classification models while using at least 6 and 10 most important predictors in undersampling and oversampling experiments, respectively. In the no-resampling experiment, the best performance (accuracy = 88.89%, sensitivity = 66.67%, specificity = 95.24%, and AUC = 0.8571) was achieved for the Naive Bayes classifier with the 12 most important features. CONCLUSION: The Naive Bayes, Ensemble, and SVM methods could be the primary choices to apply the machine learning approach in MTSS risk prediction. These predictive methods, alongside the eight common proposed predictors, might help to more accurately calculate the individual risk of developing MTSS at the point of care.

How Accurate Are the Anthropometry Equations in in Iranian Military Men in Predicting Body Composition?

BACKGROUND: The body composition varies according to different life styles (i.e. intake calories and caloric expenditure). Therefore, it is wise to record military personnel's body composition periodically and encourage those who abide to the regulations. Different methods have been introduced for body composition assessment: invasive and non-invasive. Amongst them, the Jackson and Pollock equation is most popular. OBJECTIVES: The recommended anthropometric prediction equations for assessing men's body composition were compared with dual-energy X-ray absorptiometry (DEXA) gold standard to develop a modified equation to assess body composition and obesity quantitatively among Iranian military men. PATIENTS AND METHODS: A total of 101 military men aged 23 - 52 years old with a mean age of 35.5 years were recruited and evaluated in the present study (average height, 173.9 cm and weight, 81.5 kg). The body-fat percentages of subjects were assessed both with anthropometric assessment and DEXA scan. The data obtained from these two methods were then compared using multiple regression analysis. RESULTS: The mean and standard deviation of body fat percentage of the DEXA assessment was 21.2 ± 4.3 and body fat percentage obtained from three Jackson and Pollock 3-, 4- and 7-site equations were 21.1 ± 5.8, 22.2 ± 6.0 and 20.9 ± 5.7, respectively. There was a strong correlation between these three equations and DEXA (R² = 0.98). CONCLUSIONS: The mean percentage of body fat obtained from the three equations of Jackson and Pollock was very close to that of body fat obtained from DEXA; however, we suggest using a modified Jackson-Pollock 3-site equation for volunteer military men because the 3-site equation analysis method is simpler and faster than other methods.