Comparing the standards of one metabolic equivalent of task in accurately estimating physical activity energy expenditure based on acceleration
Comparing the standards of one metabolic equivalent of task in accurately estimating physical activity energy expenditure based on acceleration
复制标题
DOI：
10.1080/02640414.2016.1221520
复制
发表时间：
2017-01-01
影响因子：
3.4
通讯作者：
Kim, In Young
中科院分区：
文献类型：
--
作者：
Kim, Dohyun;Lee, Jongshill;Kim, In Young
本研究的目的是分析静息代谢率（RMR）的标准如何影响使用加速度计的代谢当量（MET）的估计。为了研究根据活动强度对估计的影响，在3.5 ml O-2之间进行了比较。kg（-1）。min（-1），单独测量静息VO 2作为1 MET的标准。通过使用2种MET值和加速度进行五重交叉验证得出的线性回归方程估计MET;通过交叉验证、Bland和Altman图以及单因素ANOVA检验分析估计的准确性。交叉验证后RMS误差无显著差异。然而，在改良的Bland和Altman图中，基于RMR的个体估计值的平均差异高达0.5 MET，而RMR为3.5 ml O-2。kg（-1）min（-1）。最后，ANOVA检验的结果表明，在每个活动强度下，基于RMR的个体估计值与参考值之间的差异不太显著。总之，RMR标准是影响通过加速度准确估计MET的一个因素;因此，当RMR用于使用加速度计估计MET时，需要单独的规范。
The purpose of the study is to analyse how the standard of resting metabolic rate (RMR) affects estimation of the metabolic equivalent of task (MET) using an accelerometer. In order to investigate the effect on estimation according to intensity of activity, comparisons were conducted between the 3.5 ml O-2 . kg(-1) . min(-1) and individually measured resting VO2 as the standard of 1 MET. MET was estimated by linear regression equations that were derived through five-fold cross-validation using 2 types of MET values and accelerations; the accuracy of estimation was analysed through cross-validation, Bland and Altman plot, and one-way ANOVA test. There were no significant differences in the RMS error after cross-validation. However, the individual RMR-based estimations had as many as 0.5 METs of mean difference in modified Bland and Altman plots than RMR of 3.5 ml O-2 . kg(-1) min(-1). Finally, the results of an ANOVA test indicated that the individual RMR-based estimations had less significant differences between the reference and estimated values at each intensity of activity. In conclusion, the RMR standard is a factor that affects accurate estimation of METs by acceleration; therefore, RMR requires individual specification when it is used for estimation of METs using an accelerometer.