Prediction of oxygen uptake dynamics by machine learning analysis of wearable sensors during activities of daily living

Currently, oxygen uptake () is the most precise means of investigating aerobic fitness and level of physical activity; however, can only be directly measured in supervised conditions. With the advancement of new wearable sensor technologies and data processing approaches, it is possible to accurately infer work rate and predict during activities of daily living (ADL). The main objective of this study was to develop and verify the methods required to predict and investigate the dynamics during ADL. The variables derived from the wearable sensors were used to create a predictor based on a random forest method. The temporal dynamics were assessed by the mean normalized gain amplitude (MNG) obtained from frequency domain analysis. The MNG provides a means to assess aerobic fitness. The predicted during ADL was strongly correlated (r = 0.87, P < 0.001) with the measured and the prediction bias was 0.2 ml·min−1·kg−1. The MNG calculated based on predicted was strongly correlated (r = 0.71, P < 0.001) with MNG calculated based on measured data. This new technology provides an important advance in ambulatory and continuous assessment of aerobic fitness with potential for future applications such as the early detection of deterioration of physical health.

[1]  Julien Penders,et al.  Estimating Oxygen Uptake During Nonsteady-State Activities and Transitions Using Wearable Sensors , 2016, IEEE Journal of Biomedical and Health Informatics.

[2]  Linda Tapsell,et al.  Activity counts from accelerometers do not add value to energy expenditure predictions in sedentary overweight individuals during weight loss interventions. , 2011, Journal of physical activity & health.

[3]  M. Jetté,et al.  Metabolic equivalents (METS) in exercise testing, exercise prescription, and evaluation of functional capacity , 1990, Clinical cardiology.

[4]  Ciprian M. Crainiceanu,et al.  Estimating Energy Expenditure from Heart Rate in Older Adults: A Case for Calibration , 2014, PloS one.

[5]  Andrey V. Savkin,et al.  Oxygen Uptake Estimation in Humans During Exercise Using a Hammerstein Model , 2007, Annals of Biomedical Engineering.

[6]  Toru Nakamura,et al.  Multiscale Analysis of Intensive Longitudinal Biomedical Signals and Its Clinical Applications , 2016, Proceedings of the IEEE.

[7]  D. Cunningham,et al.  The influence of age and cardiorespiratory fitness on kinetics of oxygen uptake. , 1996, Canadian journal of applied physiology = Revue canadienne de physiologie appliquee.

[8]  Richard L Hughson,et al.  Aerobic system analysis based on oxygen uptake and hip acceleration during random over-ground walking activities. , 2017, American journal of physiology. Regulatory, integrative and comparative physiology.

[9]  Uwe Hoffmann,et al.  VO2 kinetics in subjects differing in aerobic capacity: investigation by spectral analysis , 2004, European Journal of Applied Physiology and Occupational Physiology.

[10]  P. Freedson,et al.  Validity of accelerometry for the assessment of moderate intensity physical activity in the field. , 2000, Medicine and science in sports and exercise.

[11]  R. Peterka.  Sensorimotor integration in human postural control. , 2002, Journal of neurophysiology.

[12]  A. Sacchetti,et al.  Interrogation of Patient Smartphone Activity Tracker to Assist Arrhythmia Management. , 2016, Annals of emergency medicine.

[13]  J. Sallis,et al.  Assessment of Physical Activity by Self-Report: Status, Limitations, and Future Directions , 2000, Research quarterly for exercise and sport.

[14]  Hung T. Nguyen,et al.  Portable sensor based dynamic estimation of human oxygen uptake via nonlinear multivariable modelling , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[15]  Rodrigo Villar,et al.  Validation of the Hexoskin wearable vest during lying, sitting, standing, and walking activities. , 2015, Applied physiology, nutrition, and metabolism = Physiologie appliquee, nutrition et metabolisme.

[16]  D. Essfeld,et al.  A model for studying the distortion of muscle oxygen uptake patterns by circulation parameters , 2004, European Journal of Applied Physiology and Occupational Physiology.

[17]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[18]  Sari A Acra,et al.  Efficiency of walking and stepping: relationship to body fatness. , 2004, Obesity research.

[19]  S A Ward,et al.  Pulmonary gas exchange dynamics and the tolerance to muscular exercise: effects of fitness and training. , 1992, The Annals of physiological anthropology = Seiri Jinruigaku Kenkyukai kaishi.

[20]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[21]  J E Cochrane,et al.  Frequency domain analysis of ventilation and gas exchange kinetics in hypoxic exercise. , 1991, Journal of applied physiology.

[22]  S. Powers,et al.  Oxygen uptake kinetics in trained athletes differing in $$\dot V_{{\text{O}}_{{\text{2max}}} }$$ , 2004, European Journal of Applied Physiology and Occupational Physiology.

[23]  T. W. Kerlin,et al.  Frequency Response Testing in Nuclear Reactors , 1975, IEEE Transactions on Nuclear Science.

[24]  K R Westerterp,et al.  Assessment of energy expenditure by recording heart rate and body acceleration. , 1989, Medicine and science in sports and exercise.

[25]  Lu Wang,et al.  Transient and steady state estimation of human oxygen uptake based on noninvasive portable sensor measurements , 2009, Medical & Biological Engineering & Computing.

[26]  J. D. Janssen,et al.  A triaxial accelerometer and portable data processing unit for the assessment of daily physical activity , 1997, IEEE Transactions on Biomedical Engineering.

[27]  Andrey V. Savkin,et al.  Estimation of Oxygen Consumption for Moderate Exercises by Using a Hammerstein Model , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[28]  John Staudenmayer,et al.  An artificial neural network to estimate physical activity energy expenditure and identify physical activity type from an accelerometer. , 2009, Journal of applied physiology.

[29]  Richard L. Hughson,et al.  Oxygen uptake kinetics: historical perspective and future directions. , 2009, Applied physiology, nutrition, and metabolism = Physiologie appliquee, nutrition et metabolisme.

[30]  Angelo M. Sabatini,et al.  Machine Learning Methods for Classifying Human Physical Activity from On-Body Accelerometers , 2010, Sensors.

[31]  Luc Vanhees,et al.  Clinical Recommendations for Cardiopulmonary Exercise Testing Data Assessment in Specific Patient Populations , 2012, Circulation.

[32]  Jon Atli Benediktsson,et al.  Multiple Classifier Systems , 2015, Lecture Notes in Computer Science.

[33]  David A. Rowe,et al.  Using Cadence to Study Free-Living Ambulatory Behaviour , 2012, Sports Medicine.

[34]  H J Green,et al.  Progressive effect of endurance training on VO2 kinetics at the onset of submaximal exercise. , 1995, Journal of applied physiology.

[35]  J. Holloszy,et al.  Faster adjustment to and recovery from submaximal exercise in the trained state. , 1980, Journal of applied physiology: respiratory, environmental and exercise physiology.

[36]  S. Ward,et al.  Influence of exercise intensity on the on‐ and off‐transient kinetics of pulmonary oxygen uptake in humans , 2001, The Journal of physiology.

[37]  Chantal Simon,et al.  Physical Activity‐Related Energy Expenditure With the RT3 and TriTrac Accelerometers in Overweight Adults , 2007, Obesity.

[38]  Aparecida Maria Catai,et al.  Relationship between oxygen consumption kinetics and BODE Index in COPD patients , 2012, International journal of chronic obstructive pulmonary disease.

[39]  Uwe Hoffmann,et al.  Dynamic linearity ofVO2 responses during aerobic exercise , 2004, European Journal of Applied Physiology and Occupational Physiology.