Research on the prediction model of energy expenditure of health Qigong Wuqinxi established by heart rate combined with accelerometer counts
Objective To establish the energy expenditure(EE)prediction models of health Qigong Wuqinxi based on heart rate combined with accelerometer counts in different body parts,so to provide a reference for monitoring EE of Wuqinxi.Methods Seventy-four healthy college students aged 18-30 were selected as the research objects.They were divided into a skilled group of 39(21 males and 18 females)and a primary group of 35(17 males and 18 females)according to their level of practice,with 55 in the model group and 19 in the validation group.When performing a set of Wuqinxi,all subjects were recorded heart rate(HR),accelerometer counts[the average counts of X-axis,Y-axis,Z-axis and vector magnitude(VM)],and EE-related indicators by wearing the Polar heart rate moni-tor,ActiGraph-GT3X+accelerometers(9 parts:waist,both arms,both wrists,both thighs,and both ankles)and CORTEX Meta Max3B-R2 portable indirect calorimeter.Based on the accurate EE mea-sured by the gas metabolism meter,linear regression models of only HR,only accelerometer counts,or HR combined accelerometer counts were established,and their accuracy was analyzed and com-pared.Results The EE of Wuqinxi was significantly correlated with gender,height,weight,muscle mass,HR-related indicators,and accelerometer counts in different parts(X-axis counts of the waist,VM values of the left thigh and counts of the X-axis and Y-axis,VM values of the right thigh and the Y-axis,Z-axis count value,left ankle VM value and Y-axis count value,right ankle VM value and Y-axis,Z-axis count value)(P<0.05).Moreover,the adjusted R2 of the prediction model of only HR,only accelerometer counts,and the both were 0.582,0.508 and 0.678,respectively(P<0.05).The correlation between the predicted and measured values ranged between 0.706 and 0.817.Accord-ing to the Bland-Altman analysis,for each model,all except one error value fell outside the 95%confidence interval.That is,all models had an excellent fitting effect and high accuracy.Among them,the combined model was of the best prediction effect:EE(kcal)=-20.089+0.279×body weight(kg)+0.243×ΔHR(exercise HR-quiet HR,bpm)+0.001×Right thigh Y-axis count value+0.181×exer-cise HR(bpm)-4.202×gender(male=0,female=1).Conclusion The EE prediction model of Wuqinxi es-tablished on HR combined with accelerometer counts has the best effect and can calculate the EE more accurately.
health Qigong Wuqinxienergy expenditureheart rateaccelerometerprediction model
NETL
NSTL
万方数据
