Deep Learning for Treadmill-Oriented Cardiorespiratory Endurance Testing and Training

Abstract

The aim of this paper was to study deep learning for treadmill-oriented cardiorespiratory endurance testing and training. This paper designs a cardiorespiratory endurance test system for the general public based on ordinary exercise bikes, which can be used to execute training programs and improve cardiorespiratory endurance levels, system design, and implementation. Through the analysis and summary of the design principle, and the design of software and hardware, the heart rate measurement, power measurement, and constant power control are realized, and the human-computer interaction software integrated into the cardiorespiratory endurance test scheme is designed. The results show that the Pearson correlation coefficient verification results of the maximum oxygen uptake VO2max of the two groups are the correlation coefficient r = 0.938, |r > 0.8, indicating that the two groups of data have a high correlation; the significance coefficient p < 0.0S, lpl <0.0S, and the accuracy and validity of the system test are verified by the comparison experiment with the gold standard equipment Monaco MONARK power car.