Development of a multi-wear-site, deep learning-based physical activity intensity classification algorithm using raw acceleration data.

BACKGROUND: Accelerometers are widely adopted in research and consumer devices as a tool to measure physical activity. However, existing algorithms used to estimate activity intensity are wear-site-specific. Non-compliance to wear instructions may lead to misspecifications. In this study, we developed deep neural network models to classify device placement and activity intensity based on raw acceleration data. Performances of these models were evaluated by making comparisons to the ground truth and results derived from existing count-based algorithms. METHODS: 54 participants (26 adults 26.9±8.7 years; 28 children, 12.1±2.3 years) completed a series of activity tasks in a laboratory with accelerometers attached to each of their hip, wrist, and chest. Their metabolic rates at rest and during activity periods were measured using the portable COSMED K5; data were then converted to metabolic equivalents, and used as the ground truth for activity intensity. Deep neutral networks using the Long Short-Term Memory approach were trained and evaluated based on raw acceleration data collected from accelerometers. Models to classify wear-site and activity intensity, respectively, were evaluated. RESULTS: The trained models correctly classified wear-sites and activity intensities over 90% of the time, which outperformed count-based algorithms (wear-site correctly specified: 83% to 85%; wear-site misspecified: 64% to 75%). When additional parameters of age, height and weight of participants were specified, the accuracy of some prediction models surpassed 95%. CONCLUSIONS: Results of the study suggest that accelerometer placement could be determined prospectively, and non-wear-site-specific algorithms had satisfactory accuracies. The performances, in terms of intensity classification, of these models also exceeded typical count-based algorithms. Without being restricted to one specific wear-site, research protocols for accelerometers wear could allow more autonomy to participants, which may in turn improve their acceptance and compliance to wear protocols, and in turn more accurate results.

Physical Activity Patterns and Correlates of 9-Month-Old Chinese Infants in the Macau Population.

Objectives Many obesity-related chronic diseases originate from unhealthy childhood habits. The aim of this study was to describe 9-month-old infants' physical activity levels and patterns and to examine the correlates. Understanding these factors is necessary for improving the effectiveness of physical activity intervention programs for infants. Methods In total, 143 infant-mother dyads from Macau, SAR China, participated in this study. Physical activity (PA) was assessed by using the Actigraph GT3X+ accelerometer and the demographic variables were collected by questionnaires. Results The most important findings were that: (1) infants had more screen time during weekdays (p = .044); (2) infants and mothers were least active at 8 a.m. (both weekdays and weekends) in the morning and most active at 7 p.m. (weekdays) and 8 p.m. (weekends) in the evening; (3) infants' PA levels significantly correlated with their mothers' PA intensities during the weekends (r = .192, p = .036), especially the mothers' lower intensities in the mornings and evenings; (4) maternal BMI predicted the PA levels of the 9-month-old infants' (R2 = .06, ß = 29.188, p = .009). Conclusions for Practice Physical activity promotion programs for infants should be time-specific starting from early infancy. This study was one of the first to examine 9-month-old infants' PA levels, patterns and correlates. The results may be helpful in improving the effectiveness of future healthy lifestyle intervention programs for infants in Macau and in the region in general.

Effects of Ving Tsun Chinese Martial Art Training on Upper Extremity Muscle Strength and Eye-Hand Coordination in Community-Dwelling Middle-Aged and Older Adults: A Pilot Study.

Objectives. To evaluate the effects of Ving Tsun (VT) martial art training on the upper extremity muscle strength and eye-hand coordination of middle-aged and older adults. Methods. This study used a nonequivalent pretest-posttest control group design. Forty-two community-dwelling healthy adults participated in the study; 24 (mean age ± SD = 68.5 ± 6.7 years) underwent VT training for 4 weeks (a supervised VT session twice a week, plus daily home practice), and 18 (mean age ± SD = 72.0 ± 6.7 years) received no VT training and acted as controls. Shoulder and elbow isometric muscle strength and eye-hand coordination were evaluated using the Lafayette Manual Muscle Test System and a computerized finger-pointing test, respectively. Results. Elbow extensor peak force increased by 13.9% (P = 0.007) in the VT group and the time to reach peak force decreased (9.9%) differentially in the VT group compared to the control group (P = 0.033). For the eye-hand coordination assessment outcomes, reaction time increased by 2.9% in the VT group and decreased by 5.3% in the control group (P = 0.002). Conclusions. Four weeks of VT training could improve elbow extensor isometric peak force and the time to reach peak force but not eye-hand coordination in community-dwelling middle-aged and older adults.

Comparison between smartphone pedometer applications and traditional pedometers for improving physical activity and body mass index in community-dwelling older adults.

[Purpose] The effectiveness of a smartphone pedometer application was compared with that of a traditional pedometer for improving the physical activity and weight status of community-dwelling older adults. [Subjects and Methods] This study had a nonequivalent pretest-posttest control group design. Ninety-seven older adults (mean age ± SD, 60.1 ± 5.5 years) joined the smartphone pedometer group and underwent a 2-week walking intervention based on a smartphone pedometer application. Fifty-four older adults (mean age ± SD, 65.3 ± 8.7 years) joined the traditional pedometer group and underwent a 2-week walking intervention based on a traditional pedometer. The participants' physical activity was evaluated using the International Physical Activity Questionnaire-Short Form, and their weight status was quantified by calculating the body mass index. The daily pedometer count was also documented. [Results] No significant time, group, or time-by-group interaction effects were found for any of the outcome variables. However, trends of improvement in physical activity and body mass index were seen only in the smartphone pedometer group. [Conclusion] A smartphone pedometer application might be more favorable than a traditional pedometer in improving physical activity and body mass index in community-dwelling older adults. However, further experimental studies are necessary to confirm the results.

Feasibility of a Chest-worn accelerometer for physical activity measurement.

OBJECTIVES: A "proof-of-concept" study to examine the feasibility of wearing an Actigraph GT3X+ at the Chest (resembling a neck pendant) for physical activity measurement. DESIGN: A convenience sample of 45 healthy adults (23 male, mean age 20.0±4.5 years) was included in data analysis. METHODS: Participants simultaneously wore three GT3X+ accelerometers, on the Waist, Wrist, and Chest and completed 8 bouts of slow (.67, .89, 1.11m/s), average (1.33, 1.56, 1.78m/s) and fast (2.00, 2.22m/s) walking on a treadmill. Paired t-test, correlations and absolute percentage errors (APE) of accelerometer output (vector magnitude, VM) were computed for the key pairs: Waist-Wrist; and Waist-Chest. RESULTS: The Wrist-site overestimated VM to a greater extent at all speeds in comparison to the Chest. Pearson's r correlations were weaker for Waist-Wrist (<.80) in comparison to the Waist-Chest (>.85). The APE's were much lower (i.e. higher agreement) for the Chest (9.23-15.5%) compared to the Wrist (19.7-54.9%). Participants also felt the Chest-site was more acceptable than the Waist-site. CONCLUSIONS: PA measurements recorded by a Chest worn GT3X+ more closely resembled PA measurements recorded at the traditional Waist site than when compared to the Wrist site. When combined with high Chest site preference, the findings of our study suggest that the Chest is a feasible site for accelerometer wear.