DETERMINATION OF MAXIMUM OXYGEN CONSUMPTION BY MACHINE LEARNING METHODS USING STEP KINEMATICS

Öz

Maximal oxygen consumption (maxVO2) is a direct indicator of aerobic capacity. For this reason, maxVO2 measurement is of great importance both in sport branches and also in clinic. However, the fact that maxVO2 measurement systems are costly has led to the need to determine different analysis methods. In this study, it was aimed to predict maxVO2 values with machine learning models using anthropometric, kinematic, heart rate and step parameters. MaxVO2 values and heart rates of 52 male athletes participating in the study at three different running speeds on the treadmill were determined and evaluated together with anthropometric and kinematic data. Age, height, body weight, heart rate, leg length, thigh length, running speed, stride frequency, stride length parameters were presented as input to the machine learning models and the calculation of the maxVO2 value was made. In addition, four different machine learning models (Linear Regression, Support Vector Machines, Decision Trees, and Gaussian Process Regression) were used and the most successful approach was examined. The Gaussian Process Regression model was able to determine the maxVO2 value with the most successful prediction (R2=0.99) and the lowest error rate (RMSE=0.012). As a result, maxVO2 values were successfully estimated in both submaximal and maximal values using basic anthropometric measurements (height, body weight, leg and thigh length), heart rate, speed and stride parameters (stride frequency and stride length) within the scope of the study.

Anahtar Kelimeler

• Aerobic capasity
• Machine learning
• Maximal Oxygen Consumption

MAKSİMUM OKSİJEN TÜKETİMİNİN ADIM KİNEMATİKLERİ KULLANILARAK MAKİNE ÖĞRENME YÖNTEMLERİYLE BELİRLENMESİ

Öz

Maksimal oksijen tüketimi (maxVO2) aerobik kapasitenin doğrudan göstergesidir. Bu sebeple hem spor branşlarında hem de klinikte maxVO2 ölçümü oldukça büyük öneme sahiptir. Ancak maxVO2 ölçüm sistemlerinin maliyetli oluşu farklı analiz yöntemlerinin belirlenmesi ihtiyacını ortaya çıkarmıştır. Bu çalışmada da antropometrik, kinematik, kalp atım hızı ve adım parametreleri kullanılarak makine öğrenme modelleri ile maxVO2 değerlerinin tahmin edilmesi amaçlanmıştır. Çalışmaya katılan 52 erkek sporcunun koşu bandında yapılan üç farklı koşu hızında maxVO2 değerleri ve kalp atım hızları belirlenmiş, antropometrik ve kinematik veriler ile birlikte değerlendirilmiştir. Yaş, boy, vücut ağırlığı, kalp atım hızı, bacak uzunluğu, uyluk uzunluğu, hız, adım frekansı, adım uzunluğu parametreleri makine öğrenme modellerine girdi olarak sunularak maxVO2 değerinin hesaplanması istenmiştir. Ayrıca dört farklı makine öğrenme modeli (lineer regresyon, destek vektör makineleri, karar ağaçları ve gauss süreç regresyonu) denenerek en başarılı yaklaşımın hangisi olduğu incelenmiştir. Gauss Süreç Regresyonu modelinin en başarılı tahmin (R2=0.99) ve en düşük hata oranı (RMSE=0.012) ile maxVO2 değerini tahmin ettiği belirlenmiştir. Sonuç olarak çalışma kapsamında temel antropometrik ölçümler (boy, vücut ağırlığı, bacak ve uyluk uzunluğu), kalp atım hızı, hız ve adım parametreleri (adım frekansı ve adım uzunluğu) kullanılarak maxVO2 değerleri hem submaksimal hem de maksimal değerlerde başarılı olarak tahmin edilmiştir.

Anahtar Kelimeler

• aerobik kapasite
• makine öğrenmesi
• maksimal oksijen tüketimi

Kaynakça

1. Abut, F., & Akay, M. F. (2015). Machine learning and statistical methods for the prediction of maximal oxygen uptake: recent advances. Medical Devices (Auckland, NZ), 8, 369.
2. Abut, F., Akay, M. F.,George, J. (2016). Developing new VO2max prediction models from maximal, submaximal and questionnaire variables using support vector machines combined with feature selection. Comput Biol Med, 79, 182-192. https://doi.org/10.1016/j.compbiomed.2016.10.018
3. Abut, F., Akay, M. F., Yildiz, I., & George, J. (2015). Performance comparison of different machine learning methods for prediction of maximal oxygen uptake from submaximal data. Proceedings of the Eighth Engineering and Technology Symposium, Ankara, Turkey,
4. Akay, M. F., Özsert, G.,George, J. (2014). Destek Vektör Makineleri Kullanilarak Submaksimal Verilerden Maksimum Oksijen Tüketiminin Tahmin Edilmesi. Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi, 16(48), 42-48.
5. Akay, M. F., Zayid, E. I. M., Aktürk, E., & George, J. D. (2011). Artificial neural network-based model for predicting VO2max from a submaximal exercise test. Expert Systems with Applications, 38(3), 2007-2010. https://doi.org/10.1016/j.eswa.2010.07.135
6. Ashfaq, A., Cronin, N., & Müller, P. (2022). Recent advances in machine learning for maximal oxygen uptake (VO2 max) prediction: A review. Informatics in Medicine Unlocked, 100863.
7. Balke, B., & Ware, R. W. (1959). An experimental study of physical fitness of Air Force personnel. U.S. Armed Forces Med J 10:675-688
8. Beltrame, T., Amelard, R., Wong, A., & Hughson, R. L. (2017). Prediction of oxygen uptake dynamics by machine learning analysis of wearable sensors during activities of daily living. Scientific reports, 7(1), 1-8.

9. Billinger, S. A., Van Swearingen, E., McClain, M., Lentz, A. A., & Good, M. B. (2012). Recumbent stepper submaximal exercise test to predict peak oxygen uptake. Medicine and Science in Sports and Exercise, 44(8), 1539.
10. Borror, A., Mazzoleni, M., Coppock, J., Jensen, B. C., Wood, W. A., Mann, B., & Battaglini, C. L. (2019). Predicting oxygen uptake responses during cycling at varied intensities using an artificial neural network. Biomedical Human Kinetics, 11(1), 60-68.
11. Bundy, M., & Leaver, A. (2012). A Guide to Sports and Injury Management E-Book. Elsevier Health Sciences.
12. Cetin, E., Hindistan, I. E., & Ozkaya, Y. G. (2018). Effect of different training methods on stride parameters in speed maintenance phase of 100-m sprint running. The Journal of Strength & Conditioning Research, 32(5), 1263-1272.
13. Chatzilazaridis, I., Panoutsakopoulos, V., & Papaiakovou, G. (2012). Stride characteristics progress in a 40-m sprinting test executed by male preadolescent, adolescent and adult athletes. Biol Exerc 8: 58–77.
14. De Ruiter C, Verdijk PWL, Werker W., Zuidema MJ, & De Haan A. (2014). Stride frequency in relation to oxygen consumption in experienced and novice runners. European Journal of Sport Science, 14(3):251-258.
15. George, J. D., Paul, S. L., Hyde, A., Bradshaw, D. I., Vehrs, P. R., Hager, R. L., & Yanowitz, F. G. (2009). Prediction of maximum oxygen uptake using both exercise and non-exercise data. Measurement in Physical Education and Exercise Science, 13(1), 1-12.
16. Gutierrez Becker, B., Klein, T., Wachinger, C., Alzheimer's Disease Neuroimaging, I., the Australian Imaging, B., ve Lifestyle flagship study of, a. (2018). Gaussian process uncertainty in age estimation as a measure of brain abnormality. Neuroimage, 175, 246-258. https://doi.org/10.1016/j.neuroimage.2018.03.075
17. Harrison, M., Brown, G.,Cochrane, L. (1980). Maximal oxygen uptake: its measurement, application, and limitations. Aviation, Space, and Environmental Medicine, 51(10), 1123-1127.
18. Heyward, V. H., & Kotarski, M. (1992). Advanced Fitness Assessment and Exercise Prescription, ed. 2. Journal of Cardiopulmonary Rehabilitation and Prevention, 12(6), 445.
19. Jalil, N. A., Hwang, H. J.,Dawi, N. M. (2019). Machines learning trends, perspectives and prospects in education sector. Proceedings of the 2019 3rd International Conference on Education and Multimedia Technology
20. Jung, A. P. (2003). The impact of resistance training on distance running performance. Sports Med, 33(7), 539-552. https://doi.org/10.2165/00007256-200333070-00005
21. Lakomy, H., ve Lakomy, J. (1993). Estimation of maximum oxygen uptake from submaximal exercise on a Concept II rowing ergometer. Journal of sports sciences, 11(3), 227-232.
22. Quinonero-Candela, J., Rasmussen, C. E. (2005). A unifying view of sparse approximate Gaussian process regression. The Journal of Machine Learning Research, 6, 1939-1959.
23. Rasmussen, C. E. (2003, February). Gaussian processes in machine learning. In Summer school on machine learning (pp. 63-71). Springer, Berlin, Heidelberg.
24. Raj, J. S., & Ananthi, J. V. (2019). Recurrent neural networks and nonlinear prediction in support vector machines. Journal of Soft Computing Paradigm (JSCP), 1(01), 33-40.
25. Richter, C., O’Reilly, M., & Delahunt, E. (2021). Machine learning in sports science: challenges and opportunities. Sports Biomechanics, 1-7.
26. Shandhi, M. M. H., Bartlett, W. H., Heller, J. A., Etemadi, M., Young, A., Plötz, T., & Inan, O. T. (2020). Estimation of instantaneous oxygen uptake during exercise and daily activities using a wearable cardio-electromechanical and environmental sensor. IEEE Journal of Biomedical and Health Informatics, 25(3), 634-646.
27. Saunders, P. U., Pyne, D. B., Telford, R. D., & Hawley, J. A. (2004). Factors affecting running economy in trained distance runners. Sports Med, 34(7), 465-485. https://doi.org/10.2165/00007256-200434070-00005
28. Silva, H. S. d., Nakamura, F. Y., Papoti, M., Da Silva, A. S., & Dos-Santos, J. W. (2021). Relationship between heart rate, oxygen consumption, and energy expenditure in futsal. Frontiers in Psychology, 2896.
29. Sinirkavak, G., Dal, U., & Çetinkaya, Ö. (2004). Elit sporcularda vücut kompozisyonu ile maksimal oksijen kapasitesi arasındaki ilişki. Cumhuriyet Üniversitesi Tıp Fakültesi Dergisi, 26, 171-176.
30. Tartaruga, L. A., Dewolf, A. H., di Prampero, P. E., Fábrica, G., Malatesta, D., Minetti, A. E., ... & Zamparo, P. (2021). Mechanical work as a (key) determinant of energy cost in human locomotion: recent findings and future directions. Experimental Physiology, 106(9), 1897-1908.
31. Uslu S., Çetin E. (2022). Farklı ağırlıklar ile yapılan squat sıçramanın makine öğrenme yöntemleri ile değerlendirilmesi, 5(1):1-12. https://doi.org/10.38021/asbid.1071466.
32. Vapnik, V. (1999). The nature of statistical learning theory. Springer Science and Business Media.
33. Williams, C. K.,Rasmussen, C. E. (2006). Gaussian processes for machine learning (Vol. 2). MIT press Cambridge, MA.
34. Xu, M., Watanachaturaporn, P., Varshney, P. K., Arora, M. K. (2005). Decision tree regression for soft classification of remote sensing data. Remote Sensing of Environment, 97(3), 322-336.
35. Yaprak, Y., Aslan, A. (2008). Üniversite Badminton Takımı Oyuncularının Kalp debisi, VO2max ve solunum fonksiyon testlerinin Karşılaştırılması. Spormetre Beden Eğitimi ve Spor Bilimleri Dergisi, 6(2), 69-74.