System analysis of the physiological and psychophysiological determinants of purposeful physical activity and prediction of its effectiveness among students of a medical university

INTRODUCTION: In the field of sports physiology, the need to study the determinants of the effectiveness of various types of physical activity is quite urgent. A variety of physiological and psychophysiological parameters as potential predictors of purposeful physical activity determines the need to search for a new algorithm for system analysis carried out using modern methods of mathematical data processing. In this regard, it is relevant to use artificial neural networks and multifactorial regression analysis in order to solve the stated tasks. AIM: To carry out a system analysis of the individual physiological and psychophysiological determinants of human physical activity in order to predict its effectiveness. MATERIALS AND METHODS: One hundred twenty young men who did not have sports grades and did not regularly attend sports clubs voluntarily participated in the study. The subjects' motivational basis of behavior, basic physical qualities, physiological and psychophysiological parameters were evaluated. Forecasting the direction of performance was carried out using the constructed models of artificial neural network technology and multifactorial regression analysis. RESULTS: Based on the statistical processing of the obtained parameters (division into clusters, rank correlation, neural network modeling, linear regression), an algorithm was created for the correct and reliable identification of the direction of the effectiveness of physical activity when the study participants realized the basic physical characteristics (strength, dexterity, endurance, speed). The study participants were divided into homogeneous clusters: ‘effective in running disciplines’ (70 boys) and ‘effective in strength disciplines’ (50 boys). The models constructed using artificial neural network technology with the involvement of various parameters, allowed identification of the determinants of the effectiveness of physical activity (ROC sensitivity: 75.7, 86.0 and 96.5%). According to the calculated parameters of the regression equation the result was predicted in high-speed quality with an accuracy of 87.9% (p ≤ 0.001), and in power quality with an accuracy of 70.8% (p ≤ 0.004). CONCLUSION: The complex of mathematical and statistical methods of analysis selected in the work can be introduced for identification and system analysis of motor activity of individual physiological and psychophysiological determinants of physical activity to predict its effectiveness in young men.