Purpose: Physical activity classifiers are typically trained on data obtained from sensors at a set orientation. Changes in this orientation (such as being on a different wrist) result in performance degradation. This work investigates a method to obtain sensor location and orientation invariance for classification of wrist-mounted accelerometry via a technique known as domain adaption. Methods: Data was gathered from 16 participants who wore accelerometers on both wrists. Physical activity classification models were created using data from each wrist and then used to predict activities when using data from the opposing wrist. Using subspace alignment domain adaption, this procedure was then repeated to align the training and testing data before the classification stage. Results: Prediction of activity when using data where the wearer’s wrist was incorrectly specified resulted in a significant (p = .01) decrease in performance of 12%. When using domain adaption this drop in performance became negligible (M difference < 1%, p = .73). Conclusion: Domain adaption is a valuable method for achieving accurate physical activity classification independent of sensor orientation in wrist-worn accelerometry.
Twaites and Hillsdon are with the Sport and Health Sciences Faculty; Everson is with the Computer Sciences Faculty; University of Exeter, Exeter, United Kingdom. Langford is with Activinsights Ltd., Kimbolton, United Kingdom.
Allen, N., Sudlow, C., Downey, P., Peakman, T., Danesh, J., Elliott, P., . . . Sprosen, T. (2012). UK Biobank: Current status and what it means for epidemiology. 1(3), 123–126. doi:10.1016/j.hlpt.2012.07.003)| false
FaircloughS.NoonanR.RowlandsA.Van HeesV.KnowlesZ. & BoddyL. (2016). Wear compliance and activity in children wearing wrist and hip mounted accelerometers. Medicine & Science in Sports & Exercise48(2) 245–253. PubMed ID: 26375253 doi:
Fairclough, S., Noonan, R., Rowlands, A., Van Hees, V., Knowles, Z., & Boddy, L. (2016). Wear compliance and activity in children wearing wrist and hip mounted accelerometers. 48(2), 245–253. PubMed ID: 26375253 doi:10.1249/MSS.0000000000000771)| false
FernandoB.HabrardA.SebbanM. & TuytelaarsT. (2013). Unsupervised visual domain adaptation using subspace alignment. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2960–2967). doi:
PaveyT.G.GilsonN.D.GomersallS.R.ClarkB. & TrostS.G. (2017). Field evaluation of a random forest activity classifier for wrist-worn accelerometer data. Journal of Science and Medicine in Sport20(1) 75–80. PubMed ID: 27372275 doi:
PaveyT.G.GomersallS.R.ClarkB.K. & BrownW.J. (2016). The validity of the GENEActiv wrist-worn accelerometer for measuring adult sedentary time in free living. Journal of Science and Medicine in Sport19(5) 395–399. PubMed ID: 25956687 doi:
StaudenmayerJ.PoberD.CrouterS.BassettD. & FreedsonP. (2009). An artificial neural network to estimate physical activity energy expenditure and identify physical activity type from an accelerometer. Journal of Applied Physiology107(4) 1300–1307. PubMed ID: 19644028 doi:
Staudenmayer, J., Pober, D., Crouter, S., Bassett, D., & Freedson, P. (2009). An artificial neural network to estimate physical activity energy expenditure and identify physical activity type from an accelerometer. 107(4), 1300–1307. PubMed ID: 19644028 doi:10.1152/japplphysiol.00465.2009)| false
YangJ.NguyenM.N.SanP.P.LiX. & KrishnaswamyS. (2015July). Deep convolutional neural networks on multichannel time series for human activity recognition. In Proceedings of the 24th International Conference on Artificial Intelligence (Vol. 15 pp. 3995–4001).
Yang, J., Nguyen, M.N., San, P.P., Li, X., & Krishnaswamy, S. (2015, July). Deep convolutional neural networks on multichannel time series for human activity recognition. In (Vol. 15, pp. 3995–4001).)| false