Skip to main content

Advertisement

SpringerLink
Log in

Exploiting sensor data in professional road cycling: personalized data-driven approach for frequent fitness monitoring

  • Published:
Data Mining and Knowledge Discovery Aims and scope Submit manuscript

Abstract

We present a personalized approach for frequent fitness monitoring in road cycling solely relying on sensor data collected during bike rides and without the need for maximal effort tests. We use competition and training data of three world-class cyclists of Team Jumbo–Visma to construct personalised heart rate models that relate the heart rate during exercise to the pedal power signal. Our model captures the non-trivial dependency between exertion and corresponding response of the heart rate, which we show can be effectively estimated by an exponential kernel. To construct the daily heart rate models that are required for day-to-day fitness estimation, we aggregate all sessions in the previous week and apply sampling. On average, the explained variance of our models is 0.86, which we demonstrate is more than twice as large as for models that ignore the temporal integration involved in the heart’s response to exercise. We show that the fitness of a cyclist can be monitored by tracking developments of parameters of our heart rate models. In particular, we monitor the decay constant of the kernel involved, and also analytically determine virtual aerobic and anaerobic thresholds. We demonstrate that our findings for the virtual anaerobic threshold on average agree with the results of exercise tests. We believe this work is an important step forward in performance optimization by opening up avenues for switching to adaptive training programs that take into account the current physiological state of an athlete.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

Notes

  1. Although this is the most common pattern for healthy persons, different relationship are also observed. For example, there could be a linear relationship for the entire range of exercise intensities or even an upward inflection at large power outputs (Hofmann et al. 1997).

References

  • Bannister EW, Calvert TW, Savage MV, Bach T (1975) A systems model of training for athletic performance. Aust J Sports Med 7:57–61

    Google Scholar 

  • Bellenger C, Fuller J, Thomson R, Davison K, Robertson E, Buckley J (2016) Monitoring athletic training status through autonomic heart rate regulation: a systematic review and meta-analysis. Sports Med 46(10):1461–1486. https://doi.org/10.1007/s40279-016-0484-2

    Article  Google Scholar 

  • Binder RK, Wonisch M, Corra U, Cohen-Solal A, Vanhees L, Saner H, Schmid J-P (2008) Methodological approach to the first and second lactate threshold in incremental cardiopulmonary exercise testing. Eur J Prev Cardiol 15(6):726–734

    Article  Google Scholar 

  • Borresen J, Lambert MI (2009) The quantification of training load, the training response and the effect on performance. Sports Med 39:779–795. https://doi.org/10.2165/11317780-000000000-00000

    Article  Google Scholar 

  • Bourdon PC, Cardinale M, Murray A, Gastin P, Kellmann M, Varley MC, Gabbett TJ, Coutts AJ, Burgess DJ, Gregson W, Cable NT (2017) Monitoring athlete training loads: consensus statement. Int J Sports Physiol Perform 12(s2):2–1612170. https://doi.org/10.1123/IJSPP.2017-0208

    Article  Google Scholar 

  • Brooke J, Hamley E (1972) The heart-rate-physical work curve analysis for the prediction of exhausting work ability. Med Sci Sports 4(1):23–26

    Google Scholar 

  • Buchheit M (2014) Monitoring training status with hr measures: do all roads lead to rome? Front Physiol 5:73. https://doi.org/10.3389/fphys.2014.00073

    Article  Google Scholar 

  • Buchheit M, Papelier Y, Laursen PB, Ahmaidi S (2007) Noninvasive assessment of cardiac parasympathetic function: postexercise heart rate recovery or heart rate variability? Am J Physiol Heart Circ Physiol 293(1):8–10

    Article  Google Scholar 

  • Bulck DV, Weghe AV, Goossens D (2021) Result-based talent identification in road cycling: discovering the next eddy merckx. Ann Oper Res. https://doi.org/10.1007/s10479-021-04280-0

    Article  Google Scholar 

  • Bunc V, Heller J, Leso J (1988) Kinetics of heart rate responses to exercise. J Sports Sci 6(1):39–48

    Article  Google Scholar 

  • Bunc V, Hofmann P, Leitner H, Gaisl G (1995) Verification of the heart rate threshold. Eur J Appl Physiol 70(3):263–9. https://doi.org/10.1007/BF00238574

    Article  Google Scholar 

  • Cheng B, Kuipers H, Snyder AC, Keizer HA, Jeukendrup A, Hesselink M (1992) A new approach for the determination of ventilatory and lactate thresholds. Int J Sports Med 13(7):518–522. https://doi.org/10.1055/s-2007-1021309

    Article  Google Scholar 

  • Chwalbinska-Moneta J, Krysztofiak H, Ziemba A, Nazar K, Kaciuba-Uściłko H (1996) Threshold increases in plasma growth hormone in relation to plasma catecholamine and blood lactate concentrations during progressive exercise in endurance-trained athletes. Eur J Appl Physiol 73(1):117–120

    Article  Google Scholar 

  • Cohen J (1992) Statistical power analysis. Curr Dir Psychol Sci 1(3):98–101. https://doi.org/10.1111/1467-8721.ep10768783

    Article  Google Scholar 

  • Conconi F, Ferrari M, Ziglio PG, Droghetti P, Codecá L (1982) Determination of the anaerobic threshold by a noninvasive field test in runners. J Appl Physiol 52(4):869–73

    Article  Google Scholar 

  • de Leeuw A-W, van der Zwaard S, van Baar R, Knobbe A (2021) Personalized machine learning approach to injury monitoring in elite volleyball players. Eur J Sport Sci, 1–10

  • de Leeuw A-W, Heijboer M, Hofmijster M, van der Zwaard S, Knobbe A (2020) Time series regression in professional road cycling. In: Appice A, Tsoumakas G, Manolopoulos Y, Matwin S (eds) Discovery science. Springer, Cham, pp 689–703

    Chapter  Google Scholar 

  • De Spiegeleer E. Predicting cycling results using machine learning

  • Demosthenous G, Kyriakou M, Vassiliades V (2022) Deep reinforcement learning for improving competitive cycling performance. Expert Syst Appl 203:117311

    Article  Google Scholar 

  • Foster C (1998) Monitoring training in athletes with reference to overtraining syndrome. Med Sci Sports Exerc 30(7):1164–1168

    Article  Google Scholar 

  • Grazzi G, Alfieri N, Borsetto C, Casoni I, Manfredini F, Mazzoni G, Conconi F (1999) The power output/heart rate relationship in cycling: test standardization and repeatability. Med Sci Sports Exerc 31:1478–83. https://doi.org/10.1097/00005768-199910000-00019

    Article  Google Scholar 

  • Halson SL (2014) Monitoring training load to understand fatigue in athletes. Sports Med 44:139–147. https://doi.org/10.1007/s40279-014-0253-z

    Article  Google Scholar 

  • Hamilton MT, González-Alonso J, Montain S, Coyle EF (1991) Fluid replacement and glucose infusion during exercise prevent cardiovascular drift. J Appl Physiol 71(3):871–7

    Article  Google Scholar 

  • Hilmkil A, Ivarsson O, Johansson M, Kuylenstierna D, van Erp T (2018) Towards machine learning on data from professional cyclists. arXiv preprint arXiv:1808.00198

  • Hofmann P, Pokan R (2010) Value of the application of the heart rate performance curve in sports. Int J Sports Physiol Perform 5(4):437–47

    Article  Google Scholar 

  • Hofmann P, Pokan R, Preidler K, Leitner H, Szolar D, Eber B, Schwaberger G (1994) Relationship between heart rate threshold, lactate turn point and myocardial function. Int J Sports Med 15:232–7. https://doi.org/10.1055/s-2007-1021052

    Article  Google Scholar 

  • Hofmann P, Pokan R, von Duvillard SP, Seibert FJ, Zweiker R, Schmid P (1997) Heart rate performance curve during incremental cycle ergometer exercise in healthy young male subjects. Med Sci Sports Exerc 29(6):762–768

    Article  Google Scholar 

  • Hofmann P, Wonisch M, Pokan R, Schwaberger G, Smekal G, Duvillard S (2005) ß1-adenoceptor mediated origin of the heart rate performance curve deflection. Med Sci Sports Exerc 37(10):1704–9

    Article  Google Scholar 

  • Hunter A, Coggan AR, McGregor S (2019) Training and racing with a power meter. VeloPress, Boulder

    Google Scholar 

  • Janssens B, Bogaert M, Maton M (2022) Predicting the next pogačar: a data analytical approach to detect young professional cycling talents. Ann Oper Res, 1–32

  • Jeukendrup A, Diemen AV (1998) Heart rate monitoring during training and competition in cyclists. J Sports Sci 16(sup1):91–99

    Article  Google Scholar 

  • Joyner MJ, Coyle EF (2008) Endurance exercise performance: the physiology of champions. J Physiol 586(1), 35–44. https://doi.org/10.1113/jphysiol.2007.143834

  • Karetnikov A (2019) Application of data-driven analytics on sport data from a professional bicycle racing team. Eindhoven University of Technology, The Netherlands

    Google Scholar 

  • Karlsson J, Jacobs I (1982) Onset of blood lactage accumulation during muscular exercise as a threshold concept I theoretical considerations. Int J Sports Med 3(4):190–201. https://doi.org/10.1055/s-2008-1026087

    Article  Google Scholar 

  • Kataoka Y, Gray P (2019) Real-time power performance prediction in tour de france. In: Brefeld U, Davis J, Van Haaren J, Zimmermann A (eds) Mach Learn Data Min Sports Anal. Springer, Cham, pp 121–130

    Google Scholar 

  • Kellmann M (2010) Preventing overtraining in athletes in high-intensity sports and stress/recovery monitoring. Scand J Med Sci Sports 20(Suppl 2):95–102. https://doi.org/10.1111/j.1600-0838.2010.01192.x

    Article  Google Scholar 

  • Kellmann M, Bertollo M, Bosquet L, Brink M, Coutts A, Duffield R, Erlacher D, Halson S, Hecksteden A, Heidari J, Kallus K, Meeusen R, Mujika I, Robazza C, Skorski S, Venter R, Beckmann J (2018) Recovery and performance in sport: consensus statement. Int J Sports Physiol Perform 13(2):240–245. https://doi.org/10.1123/ijspp.2017-0759

    Article  Google Scholar 

  • Kholkine L, Servotte T, de Leeuw A-W, Schepper TD, Hellinckx P, Verdonck T, Latré S (2021) A learn-to-rank approach for predicting road cycling race outcomes. Front Sports Active Living. https://doi.org/10.3389/fspor.2021.714107

    Article  Google Scholar 

  • Krawczyk B (2016) Learning from imbalanced data: open challenges and future directions. Progress Artif Intell 5:221–232. https://doi.org/10.1007/s13748-016-0094-0

    Article  Google Scholar 

  • Lamberts RP, Rietjens GJ, Tijdink HH, Noakes TD, Mi L (2010) Measuring submaximal performance parameters to monitor fatigue and predict cycling performance: a case study of a world-class cyclo-cross cyclist. Eur J Appl Physiol 108:183–190. https://doi.org/10.1007/s00421-009-1291-3

    Article  Google Scholar 

  • Lefever J, Berckmans D, Aerts J-M (2014) Time-variant modelling of heart rate responses to exercise intensity during road cycling. Eur J Sport Sci 14(sup1), 406–412. https://doi.org/10.1080/17461391.2012.708791. PMID: 24444235

  • Leo P, Spragg J, Podlogar T, Lawley JS, Mujika I (2021) Power profiling and the power-duration relationship in cycling: a narrative review. Eur J Appl Physiol. https://doi.org/10.1007/s00421-021-04833-y

    Article  Google Scholar 

  • Lucía A, Carvajal A, Pérez M, Boraita A (2000) Heart rate response during incremental exercise in master runners. Jpn J Physiol 50(1):155–8. https://doi.org/10.2170/jjphysiol.50.155

    Article  Google Scholar 

  • Ludwig M, Grohganz H, Asteroth A (2016) A convolution model for heart rate prediction in physical exercise. https://doi.org/10.5220/0006030901570164

  • Maier T, Schmid L, Müller B, Steiner T, Wehrlin JP (2017) Accuracy of cycling power meters against a mathematical model of treadmill cycling. Int J Sports Med 38(6):456–461. https://doi.org/10.1055/s-0043-102945

    Article  Google Scholar 

  • Mateo-March M, Moya-Ramón M, Javaloyes A, Sánchez-Muñoz C, Clemente-Suárez VJ (2022) Validity of detrended fluctuation analysis of heart rate variability to determine intensity thresholds in elite cyclists. Eur J Sport Sci. https://doi.org/10.1080/17461391.2022.2047228 (PMID: 35238695)

  • Mazzoleni M, Battaglini C, Martin K, Coffman E, Mann B (2016) Modeling and predicting heart rate dynamics across a broad range of transient exercise intensities during cycling. Sports Eng 19:117–127. https://doi.org/10.1007/s12283-015-0193-3

    Article  Google Scholar 

  • Meeusen R, Duclos M, Foster C, Fry A, Gleeson M, Nieman D, Raglin J, Rietjens G, Steinacker J, Urhausen A (2013) European college of sport science; American college of sports medicine. prevention, diagnosis, and treatment of the overtraining syndrome: joint consensus statement of the european college of sport science and the american college of sports medicine. Med Sci Sports Exer 45(1):186–205

    Article  Google Scholar 

  • Mujika I, Padilla S (2000) Detraining: loss of training-induced physiological and performance adaptations. Part i: short term insufficient training stimulus. Sports Med. 30(2):79–87. https://doi.org/10.2165/00007256-200030020-00002

    Article  Google Scholar 

  • Mujika I, Halson S, Burke LM, Balagué G, Farrow D (2018) An integrated, multifactorial approach to periodization for optimal performance in individual and team sports. Int J Sports Physiol Perform 13(5):538–561. https://doi.org/10.1123/ijspp.2018-0093

    Article  Google Scholar 

  • Mujika I, Padilla S (2000) Detraining: loss of training-induced physiological and performance adaptations. Part ii: Long term insufficient training stimulus. Sports Med 30(3), 145–54. https://doi.org/10.2165/00007256-200030030-00001

  • Nimmerichter A, Williams C, Bachl N, Eston R (2010) Evaluation of a field test to assess performance in elite cyclists. Int J Sports Med 31:160–6. https://doi.org/10.1055/s-0029-1243222

    Article  Google Scholar 

  • Ribeiro JP, Fielding RA, Hughes V, Black A, Bochese MA, Knuttgen HG (1985) Heart rate break point may coincide with the anaerobic and not the aerobic threshold. Int J Sports Med 6(4):220–4. https://doi.org/10.1055/s-2008-1025844

    Article  Google Scholar 

  • San Millán I, Gonzalez-Haro C, Sagasti M (2009) Physiological differences between road cyclists of different categories. a new approach: 733. Med Sci Sports Exercise 41:64–65. https://doi.org/10.1249/01.mss.0000353467.61975.ae

    Article  Google Scholar 

  • Steyaert M, De Bock J, Verstockt S (2022) Sensor-based performance monitoring in track cycling. In: Brefeld U, Davis J, Van Haaren J, Zimmermann A (eds) Machine learning and data mining for sports analytics. Springer, Cham, pp 167–177

    Chapter  Google Scholar 

  • Taylor K, Chapman D, Cronin J, Newton MJ, Gill N (2012) Fatigue monitoring in high performance sport: a survey of current trends. J Aust Strength Condition 20(1):12–23

    Google Scholar 

  • Thornton HR, Delaney JA, Duthie GM, Dascombe BJ (2019) Developing athlete monitoring systems in team sports: Data analysis and visualization. Int J Sports Physiol Perform 14(6):698–705. https://doi.org/10.1123/ijspp.2018-0169

    Article  Google Scholar 

  • Thorpe RT, Atkinson G, Drust B, Gregson W (2017) Monitoring fatigue status in elite team-sport athletes: Implications for practice. Int J Sports Physiol Perform 12(Suppl 2):227–234. https://doi.org/10.1123/ijspp.2016-0434

    Article  Google Scholar 

  • Valenzuela PL, Morales JS, Foster C, Lucia A, de la Villa P (2018) Is the functional threshold power a valid surrogate of the lactate threshold? Int J Sports Physiol Perform 13(10):1293–1298. https://doi.org/10.1123/ijspp.2018-0008

    Article  Google Scholar 

  • Wallace LK, Slattery KM, Coutts AJ (2009) The ecological validity and application of the session-rpe method for quantifying training loads in swimming. J Strength Condition Res 23(1):33–38

    Article  Google Scholar 

  • Wasserman K, Whipp BJ, Davis JA (1981) Respiratory physiology of exercise: metabolism, gas exchange, and ventilatory control. Int Rev Physiol 23:149–211

    Google Scholar 

Download references

Acknowledgements

The research leading to these results received funding from ZonMW under project WielerFitheid, winner of the National SportInnovator Prize 2020.

Author information

Author notes

  1. Mathieu Heijboer, Tim Verdonck, Arno Knobbe and Steven Latré have contributed equally to this work.

Authors and Affiliations

  1. University of Antwerp - imec, Sint-Pietersvliet 7, 2000, Antwerp, Belgium

    Arie-Willem de Leeuw, Tim Verdonck & Steven Latré

  2. Team Jumbo-Visma, Het Zuiderkruis 23, 5215, MV, ’s-Hertogenbosch, The Netherlands

    Mathieu Heijboer

  3. LIACS, Leiden University, Niels Bohrweg 1, 2333, CA, Leiden, The Netherlands

    Arno Knobbe

Authors
  1. Arie-Willem de Leeuw
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Mathieu Heijboer
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Tim Verdonck
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Arno Knobbe
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Steven Latré
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Arie-Willem de Leeuw.

Ethics declarations

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Ethics and reproducibility

Informed consent of the riders was obtained for using the data in this scientific study and publishing the results. The data are property of Team Jumbo-Visma and therefore cannot be shared publicly. However, peers in road cycling with access to similar power and heart rate data can reproduce all steps in our methodology with the descriptions given in the manuscript.

Additional information

Responsible editor: Ulf Brefeld and Albrecht Zimmermann.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

de Leeuw, AW., Heijboer, M., Verdonck, T. et al. Exploiting sensor data in professional road cycling: personalized data-driven approach for frequent fitness monitoring. Data Min Knowl Disc 37, 1125–1153 (2023). https://doi.org/10.1007/s10618-022-00905-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10618-022-00905-5

Keywords

Search

Navigation