Sensor-enabled Activity Class Recognition in Preschoolers: Hip versus Wrist Data

PurposePattern recognition approaches to accelerometer data processing have emerged as viable alternatives to cut-point methods. However, few studies have explored the validity of pattern recognition approaches in preschoolers, and none have compared supervised learning algorithms trained on hip and wrist data. Purpose of this study was to develop, test, and compare activity class recognition algorithms trained on hip, wrist, and combined hip and wrist accelerometer data in preschoolers. MethodsEleven children 3–6 yr of age (mean age, 4.8 ± 0.9 yr) completed 12 developmentally appropriate physical activity (PA) trials while wearing an ActiGraph GT3X+ accelerometer on the right hip and nondominant wrist. PA trials were categorized as sedentary, light activity games, moderate-to-vigorous games, walking, and running. Random forest (RF) and support vector machine (SVM) classifiers were trained using time and frequency domain features from the vector magnitude of the raw signal. Features were extracted from 15-s nonoverlapping windows. Classifier performance was evaluated using leave-one-out cross-validation. ResultsCross-validation accuracy for the hip, wrist, and combined hip and wrist RF models was 0.80 (95% confidence interval (CI), 0.79–0.82), 0.78 (95% CI, 0.77–0.80), and 0.82 (95% CI, 0.80–0.83), respectively. Accuracy for hip, wrist, and combined hip and wrist SVM models was 0.81 (95% CI, 0.80–0.83), 0.80 (95% CI, 0.79–0.80), and 0.85 (95% CI, 0.84–0.86), respectively. Recognition accuracy was consistently excellent for sedentary (>90%); moderate for light activity games, moderate-to-vigorous games, and running (69%–79%); and modest for walking (61%–71%). ConclusionsMachine learning algorithms such as RF and SVM are useful for predicting PA class from accelerometer data collected in preschool children. Although classifiers trained on hip or wrist data provided acceptable recognition accuracy, the combination of hip and wrist accelerometer delivered better performance.

[1]  Gert R. G. Lanckriet,et al.  Hip and Wrist Accelerometer Algorithms for Free-Living Behavior Classification. , 2016, Medicine and science in sports and exercise.

[2]  S. Trost,et al.  Clinical use of objective measures of physical activity , 2013, British Journal of Sports Medicine.

[3]  A. Okely,et al.  Predictive validity of three ActiGraph energy expenditure equations for children. , 2006, Medicine and science in sports and exercise.

[4]  Jung Wook Park,et al.  Child Activity Recognition Based on Cooperative Fusion Model of a Triaxial Accelerometer and a Barometric Pressure Sensor , 2013, IEEE Journal of Biomedical and Health Informatics.

[5]  T. Mercer,et al.  Comparison of accelerometry cut points for physical activity and sedentary behavior in preschool children: a validation study. , 2012, Pediatric exercise science.

[6]  Gert R. G. Lanckriet,et al.  Objective Assessment of Physical Activity: Classifiers for Public Health. , 2016, Medicine and science in sports and exercise.

[7]  Ulf Ekelund,et al.  Predictive Validity and Classification Accuracy of ActiGraph Energy Expenditure Equations and Cut-Points in Young Children , 2013, PloS one.

[8]  N. Ruch,et al.  Recognition of activities in children by two uniaxial accelerometers in free-living conditions , 2011, European Journal of Applied Physiology.

[9]  Bronwyn K. Clark,et al.  Field evaluation of a random forest activity classifier for wrist-worn accelerometer data. , 2016, Journal of science and medicine in sport.

[10]  J. R. Landis,et al.  The measurement of observer agreement for categorical data. , 1977, Biometrics.

[11]  Rohit J. Kate,et al.  Ngram time series model to predict activity type and energy cost from wrist, hip and ankle accelerometers: implications of age , 2015, Physiological measurement.

[12]  Alex V Rowlands,et al.  Wear Compliance and Activity in Children Wearing Wrist- and Hip-Mounted Accelerometers. , 2016, Medicine and science in sports and exercise.

[13]  Claire LeBlanc,et al.  Canadian 24-Hour Movement Guidelines for Children and Youth: An Integration of Physical Activity, Sedentary Behaviour, and Sleep. , 2016, Applied physiology, nutrition, and metabolism = Physiologie appliquee, nutrition et metabolisme.

[14]  Gert R. G. Lanckriet,et al.  A random forest classifier for the prediction of energy expenditure and type of physical activity from wrist and hip accelerometers , 2014, Physiological measurement.

[15]  SHAOPENG LIU,et al.  Computational methods for estimating energy expenditure in human physical activities. , 2012, Medicine and science in sports and exercise.

[16]  C. Pollak,et al.  The role of actigraphy in the study of sleep and circadian rhythms. , 2003, Sleep.

[17]  Dinesh John,et al.  Comment on "estimating activity and sedentary behavior from an accelerometer on the hip and wrist". , 2013, Medicine and science in sports and exercise.

[18]  Basel Kikhia,et al.  Optimal Placement of Accelerometers for the Detection of Everyday Activities , 2013, Sensors.

[19]  Stewart G. Trost,et al.  State of the Art Reviews: Measurement of Physical Activity in Children and Adolescents , 2007 .

[20]  M. Hagenbuchner,et al.  Energy Cost of Physical Activities and Sedentary Behaviors in Young Children. , 2016, Journal of physical activity & health.

[21]  Wei Zhao,et al.  Support vector machines classifiers of physical activities in preschoolers , 2013, Physiological reports.

[22]  Max Kuhn,et al.  Building Predictive Models in R Using the caret Package , 2008 .

[23]  A. Okely,et al.  Methodological considerations in using accelerometers to assess habitual physical activity in children aged 0-5 years. , 2009, Journal of science and medicine in sport.

[24]  Andrea Mannini,et al.  Activity recognition using a single accelerometer placed at the wrist or ankle. , 2013, Medicine and science in sports and exercise.

[25]  Olivier Dieu,et al.  Physical activity using wrist‐worn accelerometers: comparison of dominant and non‐dominant wrist , 2017, Clinical physiology and functional imaging.

[26]  T. Baranowski,et al.  Children's Activity Rating Scale (CARS): description and calibration. , 1990, Research quarterly for exercise and sport.

[27]  Stewart G Trost,et al.  Comparison of accelerometer cut points for predicting activity intensity in youth. , 2011, Medicine and science in sports and exercise.

[28]  Maurice R. Puyau,et al.  Prediction of energy expenditure and physical activity in preschoolers. , 2014, Medicine and science in sports and exercise.

[29]  Patty S. Freedson,et al.  A comprehensive evaluation of commonly used accelerometer energy expenditure and MET prediction equations , 2011, European Journal of Applied Physiology.

[30]  Jacob Cohen,et al.  Weighted kappa: Nominal scale agreement provision for scaled disagreement or partial credit. , 1968 .

[31]  Stewart G Trost,et al.  Prediction of activity type in preschool children using machine learning techniques. , 2015, Journal of science and medicine in sport.

[32]  Weng-Keen Wong,et al.  Machine learning for activity recognition: hip versus wrist data , 2014, Physiological measurement.