Optimization and Validation of an Adjustable Activity Classification Algorithm for Assessment of Physical Behavior in Elderly

Wouter Bijnens, Jos Aarts, An Stevens, Darcy Ummels, Kenneth Meijer, Wouter Bijnens, Jos Aarts, An Stevens, Darcy Ummels, Kenneth Meijer

Abstract

Due to a lack of transparency in both algorithm and validation methodology, it is difficult for researchers and clinicians to select the appropriate tracker for their application. The aim of this work is to transparently present an adjustable physical activity classification algorithm that discriminates between dynamic, standing, and sedentary behavior. By means of easily adjustable parameters, the algorithm performance can be optimized for applications using different target populations and locations for tracker wear. Concerning an elderly target population with a tracker worn on the upper leg, the algorithm is optimized and validated under simulated free-living conditions. The fixed activity protocol (FAP) is performed by 20 participants; the simulated free-living protocol (SFP) involves another 20. Data segmentation window size and amount of physical activity threshold are optimized. The sensor orientation threshold does not vary. The validation of the algorithm is performed on 10 participants who perform the FAP and on 10 participants who perform the SFP. Percentage error (PE) and absolute percentage error (APE) are used to assess the algorithm performance. Standing and sedentary behavior are classified within acceptable limits (±10% error) both under fixed and simulated free-living conditions. Dynamic behavior is within acceptable limits under fixed conditions but has some limitations under simulated free-living conditions. We propose that this approach should be adopted by developers of activity trackers to facilitate the activity tracker selection process for researchers and clinicians. Furthermore, we are convinced that the adjustable algorithm potentially could contribute to the fast realization of new applications.

Keywords: accelerometers; algorithm; elderly; physical activity; sedentary behavior; validation.

Conflict of interest statement

Maastricht Instruments B.V. is a strategic spin-off company of Maastricht University and the manufacturer of the MOX. However, Maastricht Instruments had no role in the funding, design, execution, interpretation, or writing of the current work.

Figures

Figure A1
Figure A1
The MOX Physical Activity Monitor (a) and the placement of the monitor on the upper leg (b).
Figure A2
Figure A2
Bland–Altman plots for the inter-observer reliability for (a) dynamic, (b) standing and (c) sedentary classifications.
Figure A3
Figure A3
Bland–Altman plots for (a) dynamic, (b) standing and (c) sedentary classifications with limits of agreement.
Figure 1
Figure 1
Two groups of twenty participants were random assigned to the optimization or validation group.
Figure 2
Figure 2
Graphical representation of (a) FAP with a predefined order and duration of the listed activities and (b) SFP where participants are free to choose the order and duration of the listed activities.
Figure 3
Figure 3
Schematic overview of the physical activity classification algorithm for the tracker worn on an upper leg location. The adjustable parameters are highlighted in blue.
Figure 4
Figure 4
Results of the parameter setting optimization: (a) PE and (b) APE for the evaluated data segmentation window sizes, (c) PE and (d) APE for the amount of physical activity threshold. PE and APE for dynamic are presented in black, for standing in blue and for sedentary in brown.
Figure 5
Figure 5
Percentage Error (a) and Absolute Percentage Error (b) for the validation data set. PE and APE for dynamic are presented in black, for standing in blue and for sedentary in brown.

References

    1. World Health Organization . Global Action Plan on Physical Activity 2018–2030: More Active People for a Healthier World. World Health Organization; Geneva, Switzerland: 2018.
    1. Hamari L., Kullberg T., Ruohonen J., Heinonen O.J., Díaz-Rodríguez N., Lilius J., Pakarinen A., Myllymäki A., Leppänen V., Salanterä S., et al. Physical activity among children: Objective measurements using Fitbit One((R)) and ActiGraph. BMC Res. Notes. 2017;10:161. doi: 10.1186/s13104-017-2476-1.
    1. Van Roekel E.H., Winkler E.A., Bours M.J., Lynch B.M., Willems P.J., Meijer K., Kant I., Beets G.L., Sanduleanu S., Healy G.N., et al. Associations of sedentary time and patterns of sedentary time accumulation with health-related quality of life in colorectal cancer survivors. Prev. Med. Rep. 2016;4:262–269. doi: 10.1016/j.pmedr.2016.06.022.
    1. Verlaan L., Bolink S.A.A.N., Van Laarhoven S.N., Lipperts M., Heyligers I.C., Grimm B., Senden R. Accelerometer-based Physical Activity Monitoring in Patients with Knee Osteoarthritis: Objective and Ambulatory Assessment of Actual Physical Activity During Daily Life Circumstances. Open Biomed. Eng. J. 2015;9:157. doi: 10.2174/1874120701509010157.
    1. Karel J.M., Senden R., Janssen J.E., Savelberg H.H., Grimm B., Heyligers I.C., Peeters R., Meijer K. Towards unobtrusive in vivo monitoring of patients prone to falling; Proceedings of the Annual International Conference of the IEEE on Engineering in Medicine and Biology Society (EMBC); Buenos Aires, Argentina. 31 August–4 September 2010; pp. 5018–5021.
    1. Koene S., Dirks I., Van Mierlo E., De Vries P.R., Janssen A.J., Smeitink J.A., Bergsma A., Essers H., Meijer K., De Groot I.J., et al. Domains of Daily Physical Activity in Children with Mitochondrial Disease: A 3D Accelerometry Approach. JIMD Rep. 2017;36:7–17.
    1. Meijer K., Annegarn J., Passos V.L., Savelberg H.H., Schols A.M., Wouters E.F., Spruit M.A. Characteristics of daily arm activities in patients with COPD. Eur. Respir. J. 2014;43:1631–1641. doi: 10.1183/09031936.00082513.
    1. Verwey R., van der Weegen S., Spreeuwenberg M., Tange H., van der Weijden T., de Witte L. A monitoring and feedback tool embedded in a counselling protocol to increase physical activity of patients with COPD or type 2 diabetes in primary care: Study protocol of a three-arm cluster randomised controlled trial. BMC Fam. Pract. 2014;15:93. doi: 10.1186/1471-2296-15-93.
    1. Welk G.J., Bai Y., Lee J.M., Godino J., Saint-Maurice P.F., Carr L. Standardizing Analytic Methods and Reporting in Activity Monitor Validation Studies. Med. Sci. Sports Exerc. 2019;51:1767–1780. doi: 10.1249/MSS.0000000000001966.
    1. Bunn J.A., Navalta J.W., Fountaine C.J., Reece J.D. Current State of Commercial Wearable Technology in Physical Activity Monitoring 2015–2017. Int. J. Exerc. Sci. 2018;11:503–515.
    1. Lindemann U., Zijlstra W., Aminian K., Chastin S., de Bruin E., Helbostad J., Bussmann J. Recommendations for standardizing validation procedures assessing physical activity of older persons by monitoring body postures and movements. Sensors. 2014;14:1267–1277. doi: 10.3390/s140101267.
    1. Gorman E., Hanson H.M., Yang P.H., Khan K.M., Liu-Ambrose T., Ashe M.C. Accelerometry analysis of physical activity and sedentary behavior in older adults: A systematic review and data analysis. Eur. Rev. Aging Phys. Act. 2014;11:35–49. doi: 10.1007/s11556-013-0132-x.
    1. Hernando C., Collado E.J., Panizo N., Martinez-Navarro I., Hernando B. Establishing cut-points for physical activity classification using triaxial accelerometer in middle-aged recreational marathoners. PLoS ONE. 2018;13:e0202815. doi: 10.1371/journal.pone.0202815.
    1. Howe C.A., Clevenger K., Leslie R., Ragan M. Comparison of Accelerometer-Based Cut-Points for Children’s Physical Activity: Counts vs. Steps. Children. 2018;5:105. doi: 10.3390/children5080105.
    1. Prince S.A., Reed J.L., Mark A.E., Blanchard C., Grace S., Reid R. An examination of accelerometer cut-points for quantifying physical activity in cardiac populations. Can. J. Cardiol. 2014;30:S308. doi: 10.1016/j.cjca.2014.07.547.
    1. Banos O., Galvez J.M., Damas M., Pomares H., Rojas I. Window size impact in human activity recognition. Sensors. 2014;14:6474–6499. doi: 10.3390/s140406474.
    1. Bersch S., Azzi D., Khusainov R., Achumba I., Ries J. Sensor data acquisition and processing parameters for human activity classification. Sensors. 2014;14:4239–4270. doi: 10.3390/s140304239.
    1. Bouten C.V.C., Westerterp K.R., Verduin M., Jansse J.G. A triaxial accelerometer for the assessment of daily physical activity in relation to energy expenditure; Proceedings of the 15th Annual International Conference of the IEEE Engineering in Medicine and Biology Societ; San Diego, CA, USA. 28–31 October 1993.
    1. Lipperts M., van Laarhoven S., Senden R., Heyligers I., Grimm B. Clinical validation of a body-fixed 3D accelerometer and algorithm for activity monitoring in orthopaedic patients. J. Orthop. Transl. 2017;11:19–29. doi: 10.1016/j.jot.2017.02.003.
    1. Mathie M.J., Coster A.C.F., Lovell N.H., Celler B.G. Detection of daily physical activities using a triaxial accelerometer. Med. Biol. Eng. Comput. 2003;41:296–301. doi: 10.1007/BF02348434.
    1. Preece S.J., Goulermas J.Y., Kenney L.P., Howard D., Meijer K., Crompton R. Activity identification using body-mounted sensors—A review of classification techniques. Physiol. Meas. 2009;30:R1. doi: 10.1088/0967-3334/30/4/R01.
    1. Annegarn J., Spruit M.A., Uszko-Lencer N.H., Vanbelle S., Savelberg H.H., Schols A.M., Wouters E.F., Meijer K. Objective physical activity assessment in patients with chronic organ failure: A validation study of a new single-unit activity monitor. Arch. Phys. Med. Rehabil. 2011;92:1852–1857. doi: 10.1016/j.apmr.2011.06.021.
    1. National Institute for Public Health and the Environment Beweeggedrag Bij Personen Van 4 Jaar en Ouder in 2018. [(accessed on 10 May 2018)]; Available online: .
    1. Senden R., Savelberg H.H.C.M., Grimm B., Heyligers I.C., Meijer K. Accelerometry-based gait analysis, an additional objective approach to screen subjects at risk for falling. Gait Posture. 2012;36:296–300. doi: 10.1016/j.gaitpost.2012.03.015.
    1. Berendsen B.A., Hendriks M.R., Meijer K., Plasqui G., Schaper N.C., Savelberg H.H. Which activity monitor to use? Validity, reproducibility and user friendliness of three activity monitors. BMC Public Health. 2014;14:749. doi: 10.1186/1471-2458-14-749.
    1. Maastricht Instruments MOX1. [(accessed on 1 September 2019)]; Available online: .
    1. Mesquita R., Meijer K., Pitta F., Azcuna H., Goërtz Y.M., Essers J.M., Wouters E.F., Spruit M.A. Changes in physical activity and sedentary behaviour following pulmonary rehabilitation in patients with COPD. Respir. Med. 2017;126:122–129. doi: 10.1016/j.rmed.2017.03.029.
    1. Peters D.M., Fritz S.L., Krotish D.E. Assessing the reliability and validity of a shorter walk test compared with the 10-Meter Walk Test for measurements of gait speed in healthy, older adults. J. Geriatr. Phys. Ther. 2013;36:24–30. doi: 10.1519/JPT.0b013e318248e20d.
    1. Godfrey A., Conway R., Meagher D., ÓLaighin G. Direct measurement of human movement by accelerometry. Med. Eng. Phys. 2008;30:1364–1386. doi: 10.1016/j.medengphy.2008.09.005.
    1. Dartfish Ltd. Dartfish EasyTag-Note. [(accessed on 21 December 2017)]; Available online: .
    1. Sasaki J.E., Hickey A., Staudenmayer J., John D., Kent J.A., Freedson P.S. Performance of Activity Classification Algorithms in Free-Living Older Adults. Med. Sci. Sports. Exerc. 2016;48:941. doi: 10.1249/MSS.0000000000000844.
    1. Fida B., Bernabucci I., Bibbo D., Conforto S., Schmid M. Varying behavior of different window sizes on the classification of static and dynamic physical activities from a single accelerometer. Med. Eng. Phys. 2015;37:705–711. doi: 10.1016/j.medengphy.2015.04.005.
    1. Herssens N., Verbecque E., Hallemans A., Vereeck L., Van Rompaey V., Saeys W. Do spatiotemporal parameters and gait variability differ across the lifespan of healthy adults? A systematic review. Gait Posture. 2018;64:181–190. doi: 10.1016/j.gaitpost.2018.06.012.
    1. Allahbakhshi H., Hinrichs T., Huang H., Weibel R. The Key Factors in Physical Activity Type Detection Using Real-Life Data: A Systematic Review. Front. Physiol. 2019;10:75. doi: 10.3389/fphys.2019.00075.
    1. Farrahi V., Niemelä M., Kangas M., Korpelainen R., Jämsä T. Calibration and validation of accelerometer-based activity monitors: A systematic review of machine-learning approaches. Gait Posture. 2019;68:285–299. doi: 10.1016/j.gaitpost.2018.12.003.
    1. Liu L., Wang S., Hu B., Qiong Q., Wen J., Rosenblum D.S. Learning Structures of Interval-Based Bayesian networks in Probabilistic Generative Model for Human Complex Activity Recognition. Pattern Recognit. 2018;81:545–561. doi: 10.1016/j.patcog.2018.04.022.
    1. Hildebrandt V.H., Bernaards C., Hofstetter H. Trendrapport Bewegen en Gezondheid 2000/2014. [(accessed on 14 November 2019)];2014 Available online: .
    1. Bourke A., Ihlen E., Helbostad J. Development of a gold-standard method for the identification of sedentary, light and moderate physical activities in older adults: Definitions for video annotation. J. Sci. Med. Sport. 2018;22:557–561. doi: 10.1016/j.jsams.2018.11.011.
    1. Bourke A.K., Ihlen E.A., Van de Ven P., Nelson J., Helbostad J.L. Video analysis validation of a real-time physical activity detection algorithm based on a single waist mounted tri-axial accelerometer sensor; Proceedings of the 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC); Orlando, FL, USA. 16–20 August 2016.
    1. Burton E., Hill K.D., Lautenschlager N.T., Thøgersen-Ntoumani C., Lewin G., Boyle E., Howie E. Reliability and validity of two fitness tracker devices in the laboratory and home environment for older community-dwelling people. BMC Geriatr. 2018;18:103. doi: 10.1186/s12877-018-0793-4.
    1. Ferguson T., Rowlands A.V., Olds T., Maher C. The validity of consumer-level, activity monitors in healthy adults worn in free-living conditions: A cross-sectional study. Int. J. Behav. Nutr. Phys. Act. 2015;12:42. doi: 10.1186/s12966-015-0201-9.

Source: PubMed

스폰서 및 공동 작업자

건강 상태

약물 개입

3
구독하다