Classification accuracy for the hip and wrist was 91.0% ± 3.1% and 88.4% ± 3.0%, respectively. The hip model exhibited excellent classification accuracy for sitting (91.3%), standing (95.8%), walking (95.8%), and running (96.8%); acceptable classification accuracy for lying down (88.3%) and basketball (81.9%); and modest accuracy for dance (64.1%). The wrist model exhibited excellent classification accuracy for sitting (93.0%), standing (91.7%), and walking (95.8%); acceptable classification accuracy for basketball (86.0%); and modest accuracy for running (78.8%), lying down (74.6%) and dance (69.4%).Potential Impact: Both the hip and wrist algorithms achieved acceptable classification accuracy, allowing researchers to use either placement for activity recognition.

Previous studies have demonstrated that pattern recognition approaches to accelerometer data reduction are feasible and moderately accurate in classifying activity type in children. Whether pattern recognition techniques can be used to provide valid estimates of physical activity energy expenditure in youth remains unexplored in the research literature. Purpose To develop and test artificial neural networks (ANNs) to predict physical activity (PA) type and energy expenditure (PAEE) from processed accelerometer data collected in children and adolescents. Methods 100 participants between the ages of 5 and 15 y completed 12 activity trials that were categorized into 5 PA types: sedentary, walking, running, light intensity household activities or games, and moderate-to-vigorous intensity games or sports. During each trial, participants wore an ActiGraph GT1M on the right hip and VO2 was measured using the Oxycon Mobile portable metabolic system. ANNs to predict PA type and PAEE (METs) were developed using the following features: 10th, 25th, 50th, 75th, and 90th percentiles and the lag one autocorrelation. To determine the highest time resolution achievable, features were extracted from 10, 15, 20, 30, and 60 s windows. Accuracy was assessed by calculating the percentage of windows correctly classified and root mean square error (RMSE). Results As window size increased from 10 to 60 s, accuracy for the PA type ANN increased from 81.3% to 88.4%. RMSE for the MET prediction ANN decreased from 1.1 METs to 0.9 METs. At any given window size, RMSE values for the MET prediction ANN were 30–40% lower than conventional regression-based approaches. Conclusion ANNs can be used to predict both PA type and PAEE in children and adolescents using count data from a single waist mounted accelerometer.

, and at 12 months Group 1A had on average lower calorie intake than the groups 2A (-781.8, CI-1472.7,-91.0, p¼0.008) and 3A (-885.8, CI-1599.3,-172.2, p ¼ 0.001). There was no correlation between calories intake and waist-hip ratio at baseline and reassessment with any of the questionnaires. BMI and percentage of body fat correlated with WOMAC, WOMAC pain, VAS and Lequesne scores at baseline and at reassessment. Only changes in BMI correlated weakly with one-year results of the WOMAC (r¼-0.172, p¼0.016), WOMAC pain (r¼-0.193, p¼0.007), LEQUESNE (r¼-0.197, p¼0.006) and SF-36-MCS (r¼0.160, p¼0.027); and with changes in WOMAC (r¼0.220, p¼0.002), WOMAC pain (r¼0.199, p¼0.006), VAS (r¼0.170, p¼0.018). Changes in porcentage of body fat did not correlate with any of the pain, function and quality of life changes. Conclusions: The multi-professional treatment improved percentage of body fat at one year. This improvement does not correlate with classes, telephone calls or improvements in pain, function and quality of life questionnaires (WOMAC, WOMAC pain, VAS, LEQUESNE and SF-36). Calorie intake improved and was influenced by telephone calls and classes one month apart, and does not correlate with changes in pain, function and quality of life. Changes in BMI correlated with changes in pain and function but not with quality of life.

No abstract