Abstract
Running surfaces influence energy consumption and gait parameters including swing time and stance time. This paper compares running gait cycle time, swing time and stance time, recorded on an athletics track, soft sand, and hard sand. The training and evaluation of supervised machine learning models for running surface prediction were developed using an ankle-worn inertial sensor. Models were trained using statistical features extracted from six participants using gyroscope-based stride cycles. Six different model types were trained and the performance of each model was evaluated using precision, recall, F1-score, Matthews correlation coefficient, area under the precision–recall curve and accuracy. There was a significant statistical difference in swing time and stance time across the surfaces for all participants (p < 0.05). Athlete-independent models demonstrated acceptable ability to distinguish soft sand from the two harder surfaces (≥ 0.75 mean precision, ≥ 0.90 mean recall, ≥ 0.83 mean F1-score, ≥ 0.98 mean area under the precision–recall curve across all models), but they were poor at differentiating between athletics track and hard sand. The athlete-dependent models demonstrated strong ability to classify all the surfaces (weighted average precision, recall, F1-score, Matthews correlation coefficient, area under the precision–recall curve, and overall accuracy ≥ 96%). Support vector machine models were the best in both athlete-independent and athlete-dependent methodologies. Features extracted from an ankle-worn inertial sensor can be used to classify running surface with high performance, when models are trained using features pertinent to each athlete.
Similar content being viewed by others
References
Baumgart F (2000) Stiffness-an unknown world of mechanical science? Injury 31:14–23. https://doi.org/10.1016/S0020-1383(00)80040-6
Thiel DV, Worsey MTO, Klodzinski F, Emerson N, Espinosa HG (2020) A penetrometer for quantifying the surface stiffness of sport sand surfaces. Proceedings 49:64. https://doi.org/10.3390/proceedings2020049064
Ferris DP, Louie M, Farley CT (1998) Running in the real world: adjusting leg stiffness for different surfaces. In: Proceedings of the Royal Society of London. Series B: Biological Sciences. pp 989–994
Hardin EC, Van Den Bogert AJ, Hamill J (2004) Kinematic adaptations during running: effects of footwear, surface, and duration. Med Sci Sports Exerc 36:838–844. https://doi.org/10.1249/01.MSS.0000126605.65966.40
Ferris DP, Farley CT (1997) Interaction of leg stiffness and surface stiffness during human hopping. J Appl Physiol 82:15–22. https://doi.org/10.1152/jappl.1997.82.1.15
Ferris DP, Liang K, Farley CT (1999) Runners adjust leg stiffness for their first step on a new running surface. J Biomech 32:787–794
Worsey MTO, Espinosa HG, Shepherd JB, Lewerenz J, Klodzinski FSM, Thiel DV (2020) Features observed using multiple inertial sensors for running track and hard-soft sand running: a comparison study. Proceedings 49:12. https://doi.org/10.3390/proceedings2020049012
Binnie MJ, Dawson B, Pinnington H, Landers G, Peeling P (2014) Sand training: a review of current research and practical applications. J Sports Sci 32:8–15. https://doi.org/10.1080/02640414.2013.805239
Pinnington HC, Dawson B (2001) The energy cost of running on grass compared to soft dry beach sand. J Sci Med Sport 4:416–430. https://doi.org/10.1016/S1440-2440(01)80051-7
Pinnington H, Dawson B (2001) Running economy of elite surf iron men and male runners, on soft dry beach sand and grass. Eur J Appl Physiol 86:62–70. https://doi.org/10.1007/s004210100472
Gaudino P, Gaudino C, Alberti G, Minetti AE (2013) Biomechanics and predicted energetics of sprinting on sand: hints for soccer training. J Sci Med Sport 16:271–275. https://doi.org/10.1016/j.jsams.2012.07.003
(2019) Catapult fundamentals: what can playerload tell me about athlete work? In: Catapult sports. https://www.catapultsports.com/blog/fundamentals-playerload-athlete-work. Accessed 3 Oct 2019
Camomilla V, Bergamini E, Fantozzi S, Vannozzi G (2018) Trends supporting the in-field use of wearable inertial sensors for sport performance evaluation: a systematic review. Sensors 18:873. https://doi.org/10.3390/s18030873
Espinosa HG, Shepherd JB, Thiel DV, Worsey MTO (2019) Anytime, anywhere! Inertial sensors monitor sports performance. IEEE Potentials 38:11–16. https://doi.org/10.1109/MPOT.2019.2896343
Lee JB, Mellifont RB, Burkett BJ (2010) The use of a single inertial sensor to identify stride, step, and stance durations of running gait. J Sci Med Sport 13:270–273. https://doi.org/10.1016/j.jsams.2009.01.005
Lee JB, Sutter KJ, Askew CD, Burkett BJ (2010) Identifying symmetry in running gait using a single inertial sensor. J Sci Med Sport 13:559–563. https://doi.org/10.1016/j.jsams.2009.08.004
Thiel DV, Shepherd J, Espinosa HG, Kenny M, Fischer K, Worsey M, Matsuo A, Wada T (2018) Predicting ground reaction forces in sprint running using a shank mounted inertial measurement unit. Proceedings 2:199. https://doi.org/10.3390/proceedings2060199
Nyan MN, Tay FEH, Seah KHW, Sitoh YY (2006) Classification of gait patterns in the time–frequency domain. J Biomech 39:2647–2656. https://doi.org/10.1016/j.jbiomech.2005.08.014
Lopez-Nava IH, Valentín-Coronado LM, Garcia-Constantino M, Favela J (2020) Gait activity classification on unbalanced data from inertial sensors using shallow and deep Learning. Sensors 20:4756. https://doi.org/10.3390/s20174756
Cust EE, Sweeting AJ, Ball K, Robertson S (2019) Machine and deep learning for sport-specific movement recognition: a systematic review of model development and performance. J Sports Sci 37:568–600. https://doi.org/10.1080/02640414.2018.1521769
McGrath J, Neville J, Stewart T, Cronin J (2020) Upper body activity classification using an inertial measurement unit in court and field-based sports: a systematic review. Proc Inst Mech Eng Part P. https://doi.org/10.1177/1754337120959754
Einicke GA, Sabti HA, Thiel DV, Fernandez M (2018) Maximum-entropy-rate selection of features for classifying changes in knee and ankle dynamics during running. IEEE J Biomed Health Inform 22:1097–1103. https://doi.org/10.1109/JBHI.2017.2711487
Buckley C, O’Reilly MA, Whelan D, Farrell AV, Clark L, Longo V, Gilchrist MD, Caulfield B (2017) Binary classification of running fatigue using a single inertial measurement unit. In: 2017 IEEE 14th International Conference on Wearable and Implantable Body Sensor Networks (BSN). IEEE, Eindhoven, Netherlands, pp 197–201
Ohgi Y, Kaneda K, Takakura A (2014) Sensor data mining on the cinematical characteristics of the competitive swimming. Procedia Eng 72:829–834. https://doi.org/10.1016/j.proeng.2014.06.036
Bai L, Efstratiou C, Ang CS (2016) weSport: Utilising wrist-band sensing to detect player activities in basketball games. In: 2016 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops). IEEE, Sydney, Australia, pp 1–6
O’Reilly M, Whelan D, Chanialidis C, Friel N, Delahunt E, Ward T, Caulfield B (2015) Evaluating squat performance with a single inertial measurement unit. In: 12th IEEE International Conference on Wearable and Implantable Body Sensor Networks, BSN 2015, Cambridge, MA, USA, June 9–12, 2015. pp 1–6
O’Reilly MA, Whelan DF, Ward TE, Delahunt E, Caulfield BM (2017) Classification of deadlift biomechanics with wearable inertial measurement units. J Biomech 58:155–161. https://doi.org/10.1016/j.jbiomech.2017.04.028
Schuldhaus D, Kugler P, Jensen U, Eskofier B, Schlarb H, Leible M (2012) Classification of surfaces and inclinations during outdoor running using shoe-mounted inertial sensors. In: Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012). Tsukuba, japan, pp 2258–2261
Eskofier B, Oleson M, DiBenedetto C, Hornegger J (2009) Embedded surface classification in digital sports. Pattern Recognit Lett 30:1448–1456. https://doi.org/10.1016/j.patrec.2009.08.004
Dixon PC, Schütte KH, Vanwanseele B, Jacobs JV, Dennerlein JT, Schiffman JM, Fournier P-A, Hu B (2019) Machine learning algorithms can classify outdoor terrain types during running using accelerometry data. Gait Posture 74:176–181. https://doi.org/10.1016/j.gaitpost.2019.09.005
Barrett RS, Neal RJ, Roberts LJ (1998) The dynamic loading response of surfaces encountered in beach running. J Sci Med Sport 1:1–11. https://doi.org/10.1016/S1440-2440(98)80003-0
Willwacher S, Fischer KM, Rohr E, Trudeau MB, Hamill J, Brüggemann G-P (2020) Surface stiffness and footwear affect the loading stimulus for lower extremity muscles when running. J Strength Cond Res. https://doi.org/10.1519/JSC.0000000000003410 (Publish Ahead of Print)
Hayati H, Eager D, Peham C, Qi Y (2020) Dynamic behaviour of high performance of sand surfaces used in the sports industry. Vibration 3:410–424. https://doi.org/10.3390/vibration3040026
Fraccaro P, Walsh L, Doyle J, O’Sullivan D (2014) Real-world gyroscope-based gait event detection and gait feature extraction. In: Proceedings of eTELEMED, The Sixth International Conference on eHealth, Telemedicine, and Social medicine. Barcelona, Spain, pp 1–7
McGrath D, Greene BR, O’Donovan KJ, Caulfield B (2012) Gyroscope-based assessment of temporal gait parameters during treadmill walking and running. Sports Eng 15:207–213. https://doi.org/10.1007/s12283-012-0093-8
Greene BR, McGrath D, O’Neill R, O’Donovan KJ, Burns A, Caulfield B (2010) An adaptive gyroscope-based algorithm for temporal gait analysis. Med Biol Eng Comput 48:1251–1260. https://doi.org/10.1007/s11517-010-0692-0
Pinnington HC, Lloyd DG, Besier TF, Dawson B (2005) Kinematic and electromyography analysis of submaximal differences running on a firm surface compared with soft, dry sand. Eur J Appl Physiol 94:242–253. https://doi.org/10.1007/s00421-005-1323-6
Espinosa HG, Lee J, James DA (2015) The inertial sensor: a base platform for wider adoption in sports science and applications. J Fit Res 4:13–20
Shepherd JB, Thiel DV, Espinosa HG (2017) Evaluating the use of inertial-magnetic sensors to assess fatigue in boxing during intensive training. IEEE Sens Lett 1:1–4. https://doi.org/10.1109/LSENS.2017.2689919
Worsey MTO, Espinosa HG, Shepherd JB, Thiel DV (2020) An evaluation of wearable inertial sensor configuration and supervised machine learning models for automatic punch classification in boxing. IoT 1:360–381. https://doi.org/10.3390/iot1020021
Lai A, James DA, Hayes JP, Harvey EC (2004) Semi-automatic calibration technique using six inertial frames of reference. In: Abbott D, Eshraghian K, Musca CA, Pavlidis D, Weste N (eds) Proceedings. SPIE, Perth, pp 1–13
Madgwick SOH, Harrison AJL, Vaidyanathan R (2011) Estimation of IMU and MARG orientation using a gradient descent algorithm. In: 2011 IEEE International Conference on Rehabilitation Robotics. IEEE, Zurich, pp 1–7
Wada T, Nagahara R, Gleadhill S, Ishizuka T, Ohnuma H, Ohgi Y (2020) Measurement of pelvic orientation angles during sprinting using a single inertial sensor. Proceedings 49:10. https://doi.org/10.3390/proceedings2020049010
McGrath JW, Neville J, Stewart T, Cronin J (2019) Cricket fast bowling detection in a training setting using an inertial measurement unit and machine learning. J Sports Sci 37:1220–1226. https://doi.org/10.1080/02640414.2018.1553270
Kautz T, Groh BH, Hannink J, Jensen U, Strubberg H, Eskofier BM (2017) Activity recognition in beach volleyball using a deep convolutional Neural Network: leveraging the potential of deep learning in sports. Data Min Knowl Discov 31:1678–1705. https://doi.org/10.1007/s10618-017-0495-0
Abdi H, Williams LJ (2010) Principal component analysis. Wiley Interdiscip Rev Comput Stat 2:433–459. https://doi.org/10.1002/wics.101
Jolliffe IT (2002) Principal component analysis. Springer, New York
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D (2011) Scikit-learn: machine learning in python. J Mach Learn Res 12:2825–2830
McKinney W (2010) Data structures for statistical computing in python. In: Proceedings of the 9th Python in Science Conference. Austin, Texas, pp 56–61
Worsey MTO, Espinosa HG, Shepherd JB, Thiel DV (2021) One size doesn’t fit all: supervised machine learning classification in athlete-monitoring. IEEE Sens Lett 5:1–4. https://doi.org/10.1109/LSENS.2021.3060376
McGrath J, Neville J, Stewart T, Cronin J (2020) Upper body activity classification using an inertial measurement unit in court and field-based sports: a systematic review. Proc Inst Mech Eng Part P. https://doi.org/10.1177/1754337120959754s
Caporaso T, Grazioso S (2020) IART: inertial assistant referee and trainer for race walking. Sensors 20:783. https://doi.org/10.3390/s20030783
Chicco D, Jurman G (2020) The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genomics 21:1–13. https://doi.org/10.1186/s12864-019-6413-7
Forman G, Scholz M (2010) Apples-to-apples in cross-validation studies: pitfalls in classifier performance measurement. ACM SIGKDD Explor Newsl 12:49–57
Oliveira AS, Pirscoveanu CI (2021) Implications of sample size and acquired number of steps to investigate running biomechanics. Sci Rep 11:3083. https://doi.org/10.1038/s41598-021-82876-z
American Society for Testing and Materials (2016) End use products: standard test method for impact attenuation of playing surface systems, other protective sport systems, and materials used for athletics, recreation, and play. In: Annual Book of ASTM Standards; ASTM International: West Conshohocken, Volume 15, pp. F355–F416.
Acknowledgements
This project was aided by an affiliation of the lead researcher with the Sport Performance Innovation and Knowledge Excellence (SPIKE) unit of the Queensland Academy of Sport. We thank the voluntary participants who agreed to run for this research project (Ethics approval GU 2017/587). M.T.O.W. received a PhD scholarship funding from Griffith University.
Funding
This research received no external funding from grants.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Hugo Espinosa is an Associate Editor for Sports Engineering, David Thiel serves on the Editorial Board of Sports Engineering and Jonathan Shepherd is the President of the International Sports Engineering Association (ISEA), and they were not involved in the blind peer review process of this work.
Ethical approval
Ethics approval GU 2017/587.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This article is a part of Topical Collection in Sports Engineering on Wearable Sensor Technology in Sports Monitoring, Edited by Dr. Hugo G Espinosa, Dr. Aimee Mears, Prof. Andy Stamm, Prof. Yuji Ohgi and Ms. Christine Coniglio.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Worsey, M.T.O., Espinosa, H.G., Shepherd, J.B. et al. Automatic classification of running surfaces using an ankle-worn inertial sensor. Sports Eng 24, 22 (2021). https://doi.org/10.1007/s12283-021-00359-w
Accepted:
Published:
DOI: https://doi.org/10.1007/s12283-021-00359-w