Oxygen Uptake Kinetics in Sport, Exercise and Medicine: Research and Practical Applications

Free download. Book file PDF easily for everyone and every device. You can download and read online Oxygen Uptake Kinetics in Sport, Exercise and Medicine: Research and Practical Applications file PDF Book only if you are registered here. And also you can download or read online all Book PDF file that related with Oxygen Uptake Kinetics in Sport, Exercise and Medicine: Research and Practical Applications book. Happy reading Oxygen Uptake Kinetics in Sport, Exercise and Medicine: Research and Practical Applications Bookeveryone. Download file Free Book PDF Oxygen Uptake Kinetics in Sport, Exercise and Medicine: Research and Practical Applications at Complete PDF Library. This Book have some digital formats such us :paperbook, ebook, kindle, epub, fb2 and another formats. Here is The CompletePDF Book Library. It's free to register here to get Book file PDF Oxygen Uptake Kinetics in Sport, Exercise and Medicine: Research and Practical Applications Pocket Guide.

The responses were analyzed between the fundamental frequency 0. Non-linearities might introduce misinterpretation about the aerobic system temporal dynamics study where the dynamics at higher frequencies might be also a consequence of circulatory distortions As a characteristic of the PRTS protocol 15 , the amplitudes for the even harmonics were excluded a priori due to the absence of system stimulus. This procedure eliminates the influence of the system static gain over the temporal characteristics of the system which ultimately are related to aerobic power 24 , 25 , 26 , Finally, the mean of the normalized gains MNG was used as an index of the system temporal dynamics.

Higher MNG values mean faster aerobic responses. For each participant and considering the entire group response, the predicted data were validated during the PRTS and ADL using the raw measured data as reference without 0. The r coefficient, Bland-Altman plot, confidence interval and Student t-test were used for data validation.

Publications by: Sport & Exercise Science

To further explore the predictions during ADL , the sample was also clustered into three groups according to the metabolic equivalent METS estimated from the measured. Therefore, the average of the samples within the intervals 2—3. Figure 3 displays the comparison of the measured and predicted. The data obtained during ADL are displayed in Fig. Considering the entire sample for ADL , the bias 0. The was 6.

EPOC

The relative distribution of the error is plotted in Fig. A and D : linear correlation of the measured and predicted oxygen uptake between all participants. B and E : Bland-Altman plot of the predicted and measured data. C and F : distribution of the prediction error. The was 4.

Duplicate citations

The ability of the random forest algorithm in estimate different levels of metabolic equivalent at rest resting metabolic rate, RMR and during light and moderate ADL is depicted in Fig. These data were based on the same data displayed in Fig. The proximity of the estimated METS to the equality line demonstrates that the random forest was able to dissociate between different metabolic demands.

The proposed algorithm can be used to classify activity levels between light and moderate ADL.

The group mean response for the second-by-second average during the PRTS protocol was computed and depicted in Fig. The was A linear correlation between the mean normalized gain MNG calculated from predicted and measured oxygen uptake data. B Bland-Altman plot of the data displayed in A. In agreement to the initial hypothesis, the signals obtained from the wearable sensors allowed the prediction of oxygen uptake during activities of daily living and random paced walking.

Aerobic system temporal dynamics assessed by the MNG from the predicted oxygen uptake were similar to those of oxygen uptake measured by a portable metabolic device. In addition, the random forest algorithm was able to identify physical activity levels and the resting metabolic demand. Estimating the correct measurement of physical activity level during realistic scenarios remains a challenge 30 and hence, new wearable technologies and data processing approaches are necessary. The quantification of the physical activity level usually involves the estimation of energy expenditure by indirect calorimetry.

Indirect calorimetry has also been is used to calibrate wearable sensors for a wide range of activities during steady-state, allowing the energy expenditure estimation i. However, the complexity and diversity of ADL represent a challenge for the precise physical activity estimation in a realistic scenario 9 , 32 , 33 , 34 , During randomly varying exercise intensities, assessment of the rate at which adapts to the metabolic demands is indicative of aerobic fitness 24 , 25 , Thus, the ability to predict with an adequate time resolution provides an opportunity to obtain valuable information about cardiovascular health in addition to standard estimates of energy expenditure.

Previous approaches to this problem have been restricted to studies conducted under controlled laboratory conditions 10 , 12 , In the present study, we investigated a simulated ADL protocol as well as an over-ground walking protocol PRTS that mimicked the dynamic changes in walking cadences expected during daily activities. The PRTS protocol offered an optimized stimulus for the aerobic system analysis through the study of the temporal dynamics and its prediction by a random forest machine learning regression model.

Recently, Altini et al. Their algorithm combined an activity classification method with a numerical prediction approach that predicted during dynamic phases of moderate ADL. However, the ability of the algorithm to correctly identify the dynamics was reported only as a lower error of the estimation during exercise transitions.

No further validation of the modelling parameters was carried out to explore the characterization of the aerobic adjustment dynamics with eventual health-related outcome. In addition to HR and accelerometer 9 , 10 , the acquisition of more biological data such as and BF improved the estimation during transitions and steady-state. Therefore, the integration of respiratory measurements for prediction seems to be indicated, evidencing some advantages of the smart-shirts over simpler wearable devices. As with the majority of the biological processes, and BF signals are also delayed during transitions and despite not having exactly the same dynamics as the , they have predictable relationships 38 which would contribute to a better understanding of the biological variability during transitions.

David C Poole > Compare Discount Book Prices & Save up to 90% > dravtacenchurof.cf

Studies that optimize the prediction during exercise transition with the intention to better estimate energy expenditure might be controversial. The O 2 deficit at the on-transition phase is counter-balanced by the excess of O 2 consumption during recovery 39 thus the calorie counts based on different predicted temporal dynamics should be almost similar. The energy expenditure estimation is independent on the temporal dynamics, being determined only by the correct system static gain estimation.


  1. Multinational Companies, Knowledge and Technology Transfer: Turkeys Automotive Industry in Focus!
  2. Among the Faithful;
  3. Applied Physiology, Nutrition, and Metabolism.
  4. Original Research ARTICLE!
  5. Helen Carter.
  6. Kundrecensioner;
  7. A Yogi Approach!

Therefore, in terms of calories i. The responses during transitions have been used to assess aerobic fitness in constrained settings 25 , 40 and the expansion of these approaches outside of the laboratory environment represents the possibility to track changes in aerobic fitness and physical health on a daily basis.

Testing Aerobic Capacity

The assessment of aerobic fitness by wearable sensors during unsupervised daily living routine seems very promising. Therefore, the proposed algorithm can be used in the future for aerobic fitness assessment based on predicted data obtained from wearable sensors during transitions encountered during ADL for ordinary people or patient populations, or during prescribed variations in work rate, such as athletic training.

The purpose of the current study was to predict during the most common ADL. Any studies that investigate the algorithm proposed in the current study for high intensity activities must recognize that dynamics become more complex under these conditions with the potential for nonlinear contributions. The predictor developed in this study can be applied to evaluate the aerobic system dynamics during ADL where intense activities are unlikely to occur The population tested in the current study healthy men had narrow weight and age ranges which might also restrict the use of the proposed algorithm.

Further studies are necessary to verify the reliability of the predictions in different populations.

Main navigation

It is recommended that any future study incorporate dynamic protocols such as the PRTS to evaluate the ability of the proposed algorithms to predict the dynamics during exercise transitions. In conclusion, oxygen consumption dynamics can be predicted from the fusion of data from non-intrusive wearable sensors and machine learning prediction algorithms. Longitudinal predictions of oxygen uptake can be obtained from wearables based on the validation completed in the current study for activities of daily living and random over-ground walking.

The proposed random forest ensemble predictor in conjunction with MNG can be used to investigate aerobic response during realistic activities with direct applicability for the general population. Developing the aforementioned predictive model will provide a unique opportunity for continued lifelong collections in unsupervised environments.

This new technology provides a significant advance in ambulatory and continuous assessment of energy expenditure and aerobic fitness with potential for future applications such as the early detection of deterioration of physical health. How to cite this article : Beltrame, T. Prediction of oxygen uptake dynamics by machine learning analysis of wearable sensors during activities of daily living. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Meijer, G. Assessment of energy expenditure by recording heart rate and body acceleration.

This page has permanently moved.

Whipp, B. Pulmonary gas exchange dynamics and the tolerance to muscular exercise: effects of fitness and training. Borghi-Silva, A. COPD 7 , — Guazzi, M. Circulation , — Nakamura, T. IEEE , — Rudner, J.

Mannini, A. Machine learning methods for classifying human physical activity from on-body accelerometers. Sensors 10 , — Witten, I. Data Mining: Practical machine learning tools and techniques. Elsevier, Altini, M. IEEE J. Informatics 20 , — Su, S. Estimation of oxygen consumption for moderate exercises by using a Hammerstein model.