Personalized estimating metabolic equivalents by an integration system of marker-less motion capture,portable 3D scanner and calculation of mechanical work

Abstract

This study aimed to develop a personalized exercise evaluation technology that can simultaneously and accurately estimate metabolic equivalents (METs) and postural measurement during exercise to provide personalized training content, exercise intensity, and competition time according to the individual. To this end, we developed a measurement system that integrates three technologies: marker-less motion capture, human body dimension measurement, and METs estimation based on mechanical energy. For validation, the METs obtained from the exhaled gas analyzer were used as the true value, and the accuracy was compared with that estimated by an activity meter, a standard method. The test conditions included 10 adult male participants and nine different test events with varying proportions of upper and lower limbs. The validity test results showed that the estimation accuracy was equivalent to that of an activity meter. In addition, since the proposed method estimates the posture during exercise, it is possible to analyze the causes of the differences in the METs between participants in the same test. In this study, a method for estimating METs based on mechanical energy is proposed and validated. This method considers the effect of body shape since it uses marker-less motion capture to evaluate motion. The regression coefficients necessary for MET estimation are based on data from 19 participants (male:female ratio 16:3) aged 19–33 years old, and the validity evaluation was conducted on 10 male participants aged 22–24 years old. Future studies should assess the applicability of the METs estimation to other populations with relatively different body composition rates, such as males and pregnant and non-pregnant females.