Abstract
Walking is a commonly available activity to maintain a healthy lifestyle. Accurately tracking and measuring calories expended during walking can improve user feedback and intervention measures. Inertial sensors are a promising measurement tool to achieve this purpose. An important aspect in mapping inertial sensor data to energy expenditure is the question of normalizing across physiological parameters. Common approaches such as weight scaling require validation for each new population. An alternative is to use a hierarchical approach to model subject-specific parameters at one level and cross-subject parameters connected by physiological variables at a higher level. In this paper, we evaluate an inertial sensor-based hierarchical model to measure energy expenditure across a target population. We first determine the optimal movement and physiological features set to represent data. Periodicity based features are more accurate (p < 0.1 per subject) when generalizing across populations. Weight is the most accurate parameter (p < 0.1 per subject) measured as percentage prediction error. We also compare the hierarchical model with a subject-specific regression model and weight exponent scaled models. Subject-specific models perform significantly better (p < 0.1 per subject) than weight exponent scaled models at all exponent scales whereas the hierarchical model performed worse than both. However, using an informed prior from the hierarchical model produces similar errors to using a subject-specific model with large amounts of training data (p < 0.1 per subject). The results provide evidence that hierarchical modeling is a promising technique for generalized prediction energy expenditure prediction across a target population in a clinical setting.
Similar content being viewed by others
References
6DoF IMU (2009) Inertial measurement unit datasheet. Technical report, Sparkfun Electronics, 2009. http://www.sparkfun.com/products/8454
ACSM (2010) ACSM’s guidelines for exercise testing and prescription. 8th edn, Lippincott Williams and Wilkins, Philadelphia
Ainsworth BE, Haskell WL, Whitt MC, Swartz AM, Strath SJ, O’Brien WL, Bassett DR Jr, Schmitz KH, Emplaincourt PO, Jacobs DR, Leon AS (2000) Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exer 32:S498–S516
Albinali F, Intille S, Haskell W, Rosenberger M (2010) Using wearable activity type detection to improve physical activity energy expenditure estimation. In: Proceedings of the 12th conference on ubiquitous computing, New York, ACM Press, pp 311–320
Aminian K, Najafi B (2004) Capturing human motion using body-fixed sensors: outdoor measurement and clinical applications. Comput Animat Virtual Worlds 15(2):79–94
Bao L, Intille SS (2004) Activity recognition from user-annotated acceleration data. In: Alois (ed) Pervasive. Massachusetts Institute of Technology, Springer, Berlin, pp 1–17
Bishop CM (2006a) Pattern recognition and machine learning, chapter 3.3. Springer, pp 152–156
Bishop CM (2006b) Pattern recognition and machine learning, chapter 9.4. Springer, pp 450–454
Chang C, Ansari R, Khokhar AA (2004) Efficient tracking of cyclic human motion by component motion. IEEE Signal Process Lett 11(12):941–944
Crouter SE, Clowers KG, Bassett DR (2006) A novel method for using accelerometer data to predict energy expenditure. J Appl Physiol 100(4):1324–1331
Dunn AL, Marcus BH, Kampert JB, Garcia ME, Kohl HW III, Blair SB (1999) Comparison of lifestyle and structured interventions to increase physical activity and cardiorespiratory fitness: a randomized trial. JAMA 281:327–334
Freescale (2008) 1.5–6g three axis low-g micromachined accelerometer. Technical report, 2008. http://www.freescale.com/files/sensors/doc/data_sheet/MMA7260QT.pdf
Gelfand AE, Hills SE, Racine-Poon A, Smith AFM (1990) Illustration of bayesian inference in normal data models using gibbs sampling. J Am Stat Assoc 85:972–985
Gelman A, Hill J (2007) Data analysis using regression and multilevel/hierarchical models. Cambridge University Press, Cambridge
Invensense (2006) Integrated dual-axis gyro-idg 300. Technical report, 2006. http://www.sparkfun.com/datasheets/Components/IDG-300_Datasheet.pdf
Neville AM, Ramsbottom R, Williams C (1992) Scaling physiological measurements for individuals of different body size. Eur J Appl Physiol Occup Physiol 65(2):110–117
Pearce M, Cunningham D, Donner A, Rechnitzer P, Fullerton G, Howard J (1983) Energy cost of treadmill and floor walking at self-selected paces. Eur J Appl Physiol Occup Physiol 52:115–119. ISSN 0301-5548
Rogers DM, Olson BL, Wilmore JH (1995) Scaling for the vo_2-to-body size relationship among children and adults. J Appl Physiol 79(3):958–967
Rothney MP, Neumann M, Beziat A, Chen KY (2007) An artificial neural network model of energy expenditure using nonintegrated acceleration signals. J Appl Physiol 103(4):1419–1427. http://jap.physiology.org/cgi/content/abstract/103/4/1419
Tapia M (2008) Using machine learning for real-time activity recognition and estimation of energy expenditure. PhD thesis, Massachusetts Institute of Technology, Cambridge
Troiano RP (2006) Translating accelerometer counts into energy expenditure: advancing the quest. J Appl Physiol 100(4):1107–1108
Vathsangam H, Emken A, Spruijt-Metz D, Schroeder TE, Sukhatme GS (2010) Energy estimation of treadmill walking using on-body accelerometers and gyroscopes. In: Conf Proc IEEE Eng Med Biol Soc 2010:6497–6501
Vathsangam H, Emken A, Schroeder ET, Spruijt-Metz D, Sukhatme GS (2011a) Determining energy expenditure from treadmill walking using hip-worn inertial sensors: an experimental study. IEEE Trans Biomed Eng 58(10):2804–2815. ISSN 0018-9294
Vathsangam H, Emken A, Todd Schroeder E, Spruijt-Metz D, Sukhatme GS (2011b) Towards a generalized regression model for on-body energy prediction from treadmill walking. In: Proceedings of the 5th international ICST conference on pervasive computing technologies for healthcare, pp 168–175
Warburton DER, Whitney C, Bredin SSD (2006) Health benefits of physical activity. CMAJ 174(6):801–809
Waters RL, Mulroy S (1999) The energy expenditure of normal and pathologic gait. Gait Posture 9(3):207–231
Wyndham C, van der Walt W, van Rensburg A, Rogers G, Strydom N (1971) The influence of body weight on energy expenditure during walking on a road and on a treadmill. Eur J Appl Physiol Occup Physiol 29:285–292. ISSN 0301-5548
Zakeri I, Puyau MR, Adolph AL, Vohra FA, Butte NF (2006) Normalization of energy expenditure data for differences in body mass or composition in children and adolescents. J Nutr 136:1371–1376
Acknowledgments
This work was supported in part by Qualcomm, Nokia, NSF (CCR-0120778) as part of the Center for Embedded Networked Sensing (CENS), and the USC Comprehensive NCMHD Research Center of Excellence (P60 MD 002254). Support for H. Vathsangam was provided by the USC Annenberg Doctoral Fellowship Program. The authors would like to thank David Erceg of the Division of Biokinesiology and Physical Therapy, USC, for his invaluable support and guidance.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Vathsangam, H., Emken, B.A., Schroeder, E.T. et al. Hierarchical linear models for energy prediction using inertial sensors: a comparative study for treadmill walking. J Ambient Intell Human Comput 4, 747–758 (2013). https://doi.org/10.1007/s12652-012-0150-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12652-012-0150-y