Estimating Running Speed From Wrist- or Waist-Worn Wearable Accelerometer Data: A Machine Learning Approach

in Journal for the Measurement of Physical Behaviour

Click name to view affiliation

John J. Davis IVDepartment of Kinesiology, Indiana University School of Public Health, Bloomington, IN, USA

Search for other papers by John J. Davis IV in
Current site
Google Scholar
PubMed
Close
https://orcid.org/0000-0002-0518-9637*
,
Blaise E. OedingDepartment of Kinesiology, Indiana University School of Public Health, Bloomington, IN, USA

Search for other papers by Blaise E. Oeding in
Current site
Google Scholar
PubMed
Close
, and
Allison H. GruberDepartment of Kinesiology, Indiana University School of Public Health, Bloomington, IN, USA

Search for other papers by Allison H. Gruber in
Current site
Google Scholar
PubMed
Close
https://orcid.org/0000-0002-0750-5656
Restricted access

Background: Running is a popular form of exercise, and its physiological effects are strongly modulated by speed. Accelerometry-based activity monitors are commonly used to measure physical activity in research, but no method exists to estimate running speed from only accelerometer data. Methods: Using three cohorts totaling 72 subjects performing treadmill and outdoor running, we developed linear, ridge, and gradient-boosted tree regression models to estimate running speed from raw accelerometer data from waist- or wrist-worn devices. To assess model performance in a real-world scenario, we deployed the best-performing model to data from 16 additional runners completing a 13-week training program while equipped with waist-worn accelerometers and commercially available foot pods. Results: Linear, ridge, and boosted tree models estimated speed with 12.0%, 11.6%, and 11.2% mean absolute percentage error, respectively, using waist-worn accelerometer data. Errors were greater using wrist-worn data, with linear, ridge, and boosted tree models achieving 13.8%, 14.0%, and 12.8% error. Across 663 free-living runs, speed was significantly associated with run duration (p = .009) and perceived run intensity (p = .008). Speed was nonsignificantly associated with fatigue (p = .07). Estimated speeds differed from foot pod measurements by 7.25%; associations and statistical significance were similar when speed was assessed via accelerometry versus via foot pod. Conclusion: Raw accelerometry data can be used to estimate running speed in free-living data with sufficient accuracy to detect associations with important measures of health and performance. Our approach is most useful in studies where research grade accelerometry is preferable to traditional global positioning system or foot pod-based measurements, such as in large-scale observational studies on physical activity.

Supplementary Materials

    • Supplementary Figure S1 (pdf 102 KB)
    • Supplementary Figure S2 (pdf 130 KB)
    • Supplementary Table S1 (pdf 13 KB)
    • Supplementary Table S2 (pdf 124 KB)
  • Collapse
  • Expand
  • Abbiss, C.R., & Laursen, P.B. (2008). Describing and understanding pacing strategies during athletic competition. Sports Medicine, 38(3), 239252. https://doi.org/10.2165/00007256-200838030-00004

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ainsworth, B.E., Haskell, W.L., Herrmann, S.D., Meckes, N., Bassett, D.R., Tudor-Locke, C., … Leon, A.S. (2011). 2011 Compendium of physical activities: A second update of codes and MET values. Medicine & Science in Sports & Exercise, 43(8), 15751581. https://doi.org/10.1249/MSS.0b013e31821ece12

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bergstra, J., & Bengio, Y. (2012). Random search for hyper-parameter optimization. Journal of Machine Learning Research, 13(2), 281305. https://dl.acm.org/doi/10.5555/2503308.2188395

    • Search Google Scholar
    • Export Citation
  • Borg, G. (1998). Borg’s perceived exertion and pain scales. Human Kinetics.

  • Borresen, J., & Lambert, M. (2006). Validity of self-reported training duration. International Journal of Sports Science & Coaching, 1(4), 353359. https://doi.org/10.1260/174795406779367666

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chakravarty, E.F., Hubert, H.B., Lingala, V.B., & Fries, J.F. (2008). Reduced disability and mortality among aging runners: A 21-year longitudinal study. Archives of Internal Medicine, 168(15), 16381646. https://doi.org/10.1001/archinte.168.15.1638

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.

    • Search Google Scholar
    • Export Citation
  • Chin, C.-L., & Yao, G. (2014). Convergent validity. In A.C. Michalos (Ed.), Encyclopedia of quality of life and well-being research (pp. 12751276). Springer Netherlands. https://doi.org/10.1007/978-94-007-0753-5_573

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Daussin, F.N., Zoll, J., Ponsot, E., Dufour, S.P., Doutreleau, S., Lonsdorfer, E., … Geny, B. (2008). Training at high exercise intensity promotes qualitative adaptations of mitochondrial function in human skeletal muscle. Journal of Applied Physiology, 104(5), 14361441. https://doi.org/10.1152/japplphysiol.01135.2007

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Davis, J.J., Straczkiewicz, M., Harezlak, J., & Gruber, A.H. (2021). CARL: A running recognition algorithm for free-living accelerometer data. Physiological Measurement, 42(11), Article 115001. https://doi.org/10.1088/1361-6579/ac41b8

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Davison, A.C., & Hinkley, D.V. (1997). Bootstrap methods and their application. Cambridge University Press.

  • Dideriksen, M., Soegaard, C., & Nielsen, R.O. (2016). Validity of self-reported running distance. Journal of Strength and Conditioning Research, 30(6), 15921596. https://doi.org/10.1519/JSC.0000000000001244

    • Crossref
    • Search Google Scholar
    • Export Citation
  • D’Silva, A., Bhuva, A.N., van Zalen, J., Bastiaenen, R., Abdel-Gadir, A., Jones, S., … Augusto, J. (2020). Cardiovascular remodeling experienced by real-world, unsupervised, young novice marathon runners. Frontiers in Physiology, 11, Article 232. https://doi.org/10.3389/fphys.2020.00232

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ellis, K., Kerr, J., Godbole, S., Lanckriet, G., Wing, D., & Marshall, S. (2014). A random forest classifier for the prediction of energy expenditure and type of physical activity from wrist and hip accelerometers. Physiological Measurement, 35(11), 21912203. https://doi.org/10.1088/0967-3334/35/11/2191

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Firminger, C.R., Asmussen, M.J., Cigoja, S., Fletcher, J.R., Nigg, B.M., & Edwards, W.B. (2020). Cumulative metrics of tendon load and damage vary discordantly with running speed. Medicine & Science in Sports & Exercise, 52(7), 15491556. https://doi.org/10.1249/MSS.0000000000002287

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Friedman, J., Hastie, T., & Tibshirani, R. (2001). The elements of statistical learning (vol. 1). Springer Series in Statistics New York.

    • Search Google Scholar
    • Export Citation
  • Halilaj, E., Rajagopal, A., Fiterau, M., Hicks, J.L., Hastie, T.J., & Delp, S.L. (2018). Machine learning in human movement biomechanics: Best practices, common pitfalls, and new opportunities. Journal of Biomechanics, 81, 111. https://doi.org/10.1016/j.jbiomech.2018.09.009

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hastie, T.J., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning: Data mining, inference, and prediction. Springer.

  • Hernandez-Boussard, T., Bozkurt, S., Ioannidis, J.P., & Shah, N.H. (2020). MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care. Journal of the American Medical Informatics Association, 27(12), 20112015. https://doi.org/10.1093/jamia/ocaa088

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hestness, J., Narang, S., Ardalani, N., Diamos, G., Jun, H., Kianinejad, H., … Zhou, Y. (2017). Deep learning scaling is predictable, empirically. arXiv preprint arXiv:1712.00409.

    • Search Google Scholar
    • Export Citation
  • Hunter, J.G., Miller, R.H., & Suydam, S. (2017). Accuracy of a shoe-worn device to measure running mechanics. Paper Presented at the 41st Annual Meeting of the American Society of Biomechanics.

    • Search Google Scholar
    • Export Citation
  • James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning (vol. 112). Springer.

  • John, D., Miller, R., Kozey-Keadle, S., Caldwell, G., & Freedson, P. (2012). Biomechanical examination of the ‘plateau phenomenon’ in ActiGraph vertical activity counts. Physiological Measurement, 33(2), 219230. https://doi.org/10.1088/0967-3334/33/2/219

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Joseph, V.R. (2022). Optimal ratio for data splitting. Statistical Analysis and Data Mining, 15(4), 531538. https://doi.org/10.1002/sam.11583

  • Kohn, T., Essén‐Gustavsson, B., & Myburgh, K. (2011). Specific muscle adaptations in type II fibers after high‐intensity interval training of well‐trained runners. Scandinavian Journal of Medicine & Science in Sports, 21(6), 765772. https://doi.org/10.1111/j.1600-0838.2010.01136.x

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lau, P.W., Wong, D.P., Ngo, J.K., Liang, Y., Kim, C., & Kim, H. (2015). Effects of high-intensity intermittent running exercise in overweight children. European Journal of Sport Science, 15(2), 182190. https://doi.org/10.1080/17461391.2014.933880

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lee, K.A., Hicks, G., & Nino-Murcia, G. (1991). Validity and reliability of a scale to assess fatigue. Psychiatry Research, 36(3), 291298. https://doi.org/10.1016/0165-1781(91)90027-M

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mannini, A., Rosenberger, M., Haskell, W.L., Sabatini, A.M., & Intille, S.S. (2017). Activity recognition in youth using single accelerometer placed at wrist or ankle. Medicine & Science in Sports & Exercise, 49(4), 801812. https://doi.org/10.1249/MSS.0000000000001144

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McNair, D., Lorr, M., & Droppleman, L. (1981). Profile of mood states questionnaire. Educational and Industrial Testing Service.

  • Nielsen, R.O., Cederholm, P., Buist, I., Sørensen, H., Lind, M., & Rasmussen, S. (2013). Can GPS be used to detect deleterious progression in training volume among runners? Journal of Strength and Conditioning Research, 27(6), 14711478. https://doi.org/10.1519/JSC.0b013e3182711e3c

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Parker, R.A., Weir, C.J., Rubio, N., Rabinovich, R., Pinnock, H., Hanley, J., … MacNee, W. (2016). Application of mixed effects limits of agreement in the presence of multiple sources of variability: Exemplar from the comparison of several devices to measure respiratory rate in COPD patients. PLoS One, 11(12), e0168321. https://doi.org/10.1371/journal.pone.0168321

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., … Dubourg, V. (2011). Scikit-learn: Machine learning in Python. The Journal of Machine Learning Research, 12, 28252830. https://dl.acm.org/doi/10.5555/1953048.2078195

    • Search Google Scholar
    • Export Citation
  • Picard, R.R., & Berk, K.N. (1990). Data splitting. The American Statistician, 44(2), 140147. https://doi.org/10.1080/00031305.1990.10475704

    • Search Google Scholar
    • Export Citation
  • Piedmont, R.L. (2014). Criterion validity. In A.C. Michalos (Ed.), Encyclopedia of quality of life and well-being research (p. 1348). Springer Netherlands. https://doi.org/10.1007/978-94-007-0753-5_618

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pierce, J., & Kolden, C.A. (2015). The hilliness of US cities. Geographical Review, 105(4), 581600. https://doi.org/10.1111/j.1931-0846.2015.12099.x

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & Team, R.C. (2007). Linear and nonlinear mixed effects models. R Package Version, 3(57), 189.

    • Search Google Scholar
    • Export Citation
  • Potter, M.V., Ojeda, L.V., Perkins, N.C., & Cain, S.M. (2019). Effect of IMU design on IMU-derived stride metrics for running. Sensors, 19(11), Article 2601. https://doi.org/10.3390/s19112601

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raghu, M., & Schmidt, E. (2020). A survey of deep learning for scientific discovery. arXiv preprint arXiv:2003.11755.

  • Siontis, G.C., Tzoulaki, I., Castaldi, P.J., & Ioannidis, J.P. (2015). External validation of new risk prediction models is infrequent and reveals worse prognostic discrimination. Journal of Clinical Epidemiology, 68(1), 2534. https://doi.org/10.1016/j.jclinepi.2014.09.007

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Stoltenberg, B.E., Miller, E.M., Dolbeer, J.A., Pickens, B.B., & Goss, D.L. (2019). Validity of an instrumented sock and on-shoe sensor to provide biometric feedback to runners. Footwear Science, 11(3), 147152. https://doi.org/10.1080/19424280.2019.1614098

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sudlow, C., Gallacher, J., Allen, N., Beral, V., Burton, P., Danesh, J., … Landray, M. (2015). UK Biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Medicine, 12(3), e1001779. https://doi.org/10.1371/journal.pmed.1001779

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Van Buuren, S. (2018). Flexible imputation of missing data. Chapman and Hall/CRC.

  • Willetts, M., Hollowell, S., Aslett, L., Holmes, C., & Doherty, A. (2018). Statistical machine learning of sleep and physical activity phenotypes from sensor data in 96,220 UK Biobank participants. Scientific Reports, 8(1), 110. https://doi.org/10.1038/s41598-018-26174-1

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wood, S.N. (2011). Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 73(1), 336. https://doi.org/10.1111/j.1467-9868.2010.00749.x

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Woods, R.A. (2017). Spotlight on statistics: Sports and exercise. United States Bureau of Labor Statistics. Retrieved January 21, 2021 from https://www.bls.gov/spotlight/2017/sports-and-exercise/home.htm

    • Search Google Scholar
    • Export Citation
  • Yentes, J.M., Hunt, N., Schmid, K.K., Kaipust, J.P., McGrath, D., & Stergiou, N. (2013). The appropriate use of approximate entropy and sample entropy with short data sets. Annals of Biomedical Engineering, 41(2), 349365. https://doi.org/10.1007/s10439-012-0668-3

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, S., Rowlands, A.V., Murray, P., & Hurst, T.L. (2012). Physical activity classification using the GENEA wrist-worn accelerometer. Medicine & Science in Sports & Exercise, 44(4), 742748. https://doi.org/10.1249/MSS.0b013e31823bf95c

    • Crossref
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 623 623 96
Full Text Views 3 3 0
PDF Downloads 4 4 1