VO2max Prediction Models Research Articles

External validation of VO2max prediction models based on recreational and elite endurance athletes.

In recent years, numerous prognostic models have been developed to predict VO2max. Nevertheless, their accuracy in endurance athletes (EA) stays mostly unvalidated. This study aimed to compare predicted VO2max (pVO2max) with directly measured VO2max by assessing the transferability of the currently available prediction models based on their R2, calibration-in-the-large, and calibration slope. 5,260 healthy adult EA underwent a maximal exertion cardiopulmonary exercise test (CPET) (84.76% male; age 34.6±9.5 yrs.; VO2max 52.97±7.39 mL·min-1·kg-1, BMI 23.59±2.73 kg·m-2). 13 models have been selected to establish pVO2max. Participants were classified into four endurance subgroups (high-, recreational-, low- trained, and "transition") and four age subgroups (18-30, 31-45, 46-60, and ≥61 yrs.). Validation was performed according to TRIPOD guidelines. pVO2max was low-to-moderately associated with direct CPET measurements (p>0.05). Models with the highest accuracy were for males on a cycle ergometer (CE) (Kokkinos R2 = 0.64), females on CE (Kokkinos R2 = 0.65), males on a treadmill (TE) (Wasserman R2 = 0.26), females on TE (Wasserman R2 = 0.30). However, selected models underestimated pVO2max for younger and higher trained EA and overestimated for older and lower trained EA. All equations demonstrated merely moderate accuracy and should only be used as a supplemental method for physicians to estimate CRF in EA. It is necessary to derive new models on EA populations to include routinely in clinical practice and sports diagnostic.

Contribution of Body Mass Index Stratification for the Prediction of Maximal Oxygen Uptake.

The purpose of this study was to investigate whether modeling within separate body mass index (BMI) stratifications improves the accuracy of maximal oxygen uptake (VO2max) prediction compared to a model developed regardless of adults' BMIs. A total of 250 Taiwanese adults (total group, TOG) aged 22-64 years participated in this study, and were stratified into a normal group (NOG: 135), an overweight group (OVG: 69), and an obesity group (OBG: 46), according to the BMI classification recommended by the Taiwan Ministry of Health and Welfare. VO2max was directly measured on an electromagnetic bicycle ergometer. Using the participant's heart rate in the 3-min incremental step-in-place test and demographic parameters, VO2max prediction models established for four groups were TOG model, NOG model, OVG model, and OBG model, respectively. Compared with the TOG model, the OVG and OBG models had higher coefficients of determination and lower standard error of estimates (SEEs), or %SEEs. The validities of the NOG (r = 0.780), OVG (r = 0.776), and OBG (r = 0.791) models for BMI subgroups increased by 1.79%, 4.64%, and 8.22% respectively, and the reliabilities (NOG model: ICC = 0.755; OVG model: ICC = 0.765; OBG model: ICC = 0.779) increased by 3.18%, 3.27%, and 9.63%, respectively. These results suggested using separate models established in BMI stratifications can effectively improve the prediction of VO2max.

Workers\u2019 physical activity data contribute to estimating maximal oxygen consumption: a questionnaire study to concurrently assess workers\u2019 sedentary behavior and cardiorespiratory fitness

BackgroundSedentary behavior (SB) and cardiorespiratory fitness (CRF) are important issues in occupational health. Developing a questionnaire to concurrently assess workers’ SB and CRF could fundamentally improve epidemiological research. The Worker’s Living Activity-time Questionnaire (WLAQ) was developed previously to assess workers’ sitting time. WLAQ can be modified to evaluate workers’ CRF if additional physical activity (PA) data such as PA frequency, duration, and intensity are collected.MethodsA total of 198 working adults (93 women and 105 men; age, 30–60 years) completed anthropometric measurements, a treadmill exercise test for measuring maximal oxygen consumption (VO2max), and modified WLAQ (m-WLAQ, which included questions about PA data additional to the original questions). Multiple regression analyses were performed to develop prediction equations for VO2max. The generated models were cross-validated using the predicted residual error sum of squares method. Among the participants, the data of 97 participants who completed m-WLAQ twice after a 1-week interval were used to calculate intraclass correlation coefficient (ICC) for the test–retest reliability analyses.ResultsAge (r = − 0.29), sex (r = 0.48), body mass index (BMI, r = − 0.20), total sitting time (r = − 0.15), and PA score (total points for PA data, r = 0.47) were significantly correlated with VO2max. The models that included age, sex, and BMI accounted for 43% of the variance in measured VO2max [standard error of the estimate (SEE) = 5.04 ml·kg− 1·min− 1]. These percentages increased to 59% when the PA score was included in the models (SEE = 4.29 ml·kg− 1·min− 1). Cross-validation analyses demonstrated good stability of the VO2max prediction models, while systematic underestimation and overestimation of VO2max were observed in individuals with high and low fitness, respectively. The ICC of the PA score was 0.87 (0.82–0.91), indicating excellent reliability.ConclusionsThe PA score obtained using m-WLAQ, rather than sitting time, correlated well with measured VO2max. The equation model that included the PA score as well as age, sex, and BMI had a favorable validity for estimating VO2max. Thus, m-WLAQ can be a useful questionnaire to concurrently assess workers’ SB and CRF, which makes it a reasonable resource for future epidemiological surveys on occupational health.

A robust ensemble feature selector based on rank aggregation for developing new VO2max prediction models using support vector machines

This paper proposes a new ensemble feature selector, called the majority voting feature selector (MVFS), for developing new maximal oxygen uptake (VO2max) prediction models using a support vector machine (SVM). The approach is based on rank aggregation, which meaningfully utilizes the correlation among the relevance ranks of predictor variables given by three state-of-the-art feature selectors: Relief-F, minimum redundancy maximum relevance (mRMR), and maximum likelihood feature selection (MLFS). By applying the SVM combined with MVFS on a self-created dataset containing maximal and submaximal exercise data from 185 college students, several new hybrid (VO2max) prediction models have been created. To compare the performance of the proposed ensemble approach on prediction of (VO2max), SVM-based models with individual combinations of Relief-F, mRMR, and MLFS as well as with other alternative ensemble feature selectors from the literature have also been developed. The results reveal that MVFS outperforms other individual and ensemble feature selectors and yields up to 8.76 % increment and 11.15 % decrement rates in multiple correlation coefficients (Rs) and root mean square errors {RMSE}s), respectively. Furthermore, in addition to reconfirming the relevance of sex, age, and maximal heart rate in predicting VO2max, which were previously reported in the literature, it is revealed that submaximal heart rates and exercise times at 1.5-mile distance are two further discriminative predictors of (VO2max). The results have also been compared to those obtained by a general regression neural network and single decision tree combined with MVFS, and it is shown that the SVM exhibits much better performance than other methods for prediction of (VO2max).

Establishment of a Non-Exercise Questionnaire Using Physical Activity Exertion Perception to Predict Maximal Oxygen Uptake in Adults

Maximal oxygen uptake (VO2max) is an important diagnostic variable for health and fitness status. ACSM has published several sub-maximal VO2 max prediction equations. Non-exercise VO2max prediction equations currently exist, often requiring variables obtained by a fitness professional. The equation published by Uth et. al (USOP) is often used to predict VO2max from age and RHR measures. No validated models incorporate perceived exertion of physical activities. Non-exercise predictions of VO2max may provide a quick, valuable alternative to sub-maximal assessments for healthcare providers in order to stratify disease risk and prescribe aerobic exercise. PURPOSE: The purpose of the study was to explore the efficacy of a non-exercise VO2max assessment using a combination of perception of activities, body mass index (BMI), gender, and age. METHODS: Twenty-seven subjects (ages 19-49) performed a maximal graded exercise test (GXT) and the Forestry Step Test (FST). Subject anthropometrics were assessed and all subjects completed a questionnaire in which perceived exertion was estimated for 15 well known physical activities. VO2max estimations were derived from each of the activity items by dividing the metabolic equivalent by the percentage of perceived exertion. The USOP prediction of VO2max used both RHR and the Tanaka MHR equation. Data was entered into a statistical software package where five activity items with the highest correlation to GXT VO2max were used in a linear regression model to create a prediction model (PEQ). This was then compared to both the FST and USOP prediction equations. RESULTS: Mean GXT VO max were 48.50 ± 1.74 ml/kg/min-1. PEQ correlated most strongly with GXT (F(8,18)=6.159, r=0.856, p=.001), followed by FST (F(1,25)=28.635, r=.0731, p<0.001) and USOP (F(1,25)=8.575, r=0.505, p=0.007). A model using age, gender, and BMI alone yielded a Pearson correlation of 0.791. CONCLUSION: The inclusion of exertion perception in VO2max prediction models may strengthen the validity of non-exercise estimations. Future research should elucidate the most predictive activity items across populations. Given their ability to be self-administered, VO2max prediction surveys can provide valuable information to large populations where traditional evaluation methods are impractical to perform.

New prediction models for the maximal oxygen uptake of collegeaged students using non-exercise data

Maximal oxygen uptake (VO2max) refers to the maximal amount of oxygen that an individual can utilize during intense or maximal exercise. VO2max plays a significant role in sport science, education and research. The direct measurement of VO2max is time-consuming, requires expensive laboratory equipment and trained staff. Because of these disadvantages of direct measurement, numerous VO2max prediction models for a variety of subject groups have been developed. The purpose of this study is to develop new Multiple Linear Regression based on VO2max prediction models for Turkish college students by using physiological and questionnaire variables. The dataset includes the data of 62 (28 females and 34 males) students, ranging in age from 18 to 27 years, from the College of Physical Education and Sports Science at Gazi University. Seven different models consisting of the predictor variables gender, age, weight, height, Perceived Functional Ability scores (PFA1 and PFA-2), and Physical Activity Rating score (PA-R) have been used to predict VO2max. The performance of the prediction models has been evaluated by calculating their standard error of estimates (SEE’s) and multiple correlation coefficients (R’s). The prediction model including Gender, Age, Height, Weight, PFA-1 and PAR yields the lowest SEE with 5.14 mL.kg-1.min-1 and highest R with 0.93. It can be concluded that in situations where it is difficult to measure VO2max, the given model with MLR equation can be used to predict the VO2max of college students with acceptable error rates. Keywords: Maximum oxygen uptake, machine learning, multiple linear regression.

Developing new VO2max prediction models from maximal, submaximal and questionnaire variables using support vector machines combined with feature selection

Maximal oxygen uptake (VO2max) is an essential part of health and physical fitness, and refers to the highest rate of oxygen consumption an individual can attain during exhaustive exercise. In this study, for the first time in the literature, we combine the triple of maximal, submaximal and questionnaire variables to propose new VO2max prediction models using Support Vector Machines (SVM’s) combined with the Relief-F feature selector to predict and reveal the distinct predictors of VO2max. For comparison purposes, hybrid models based on double combinations of maximal, submaximal and questionnaire variables have also been developed. By utilizing 10-fold cross-validation, the performance of the models has been calculated using multiple correlation coefficient (R) and root mean square error (RMSE). The results show that the best values of R and RMSE, with 0.94 and 2.92mLkg−1min−1 respectively, have been obtained by combining the triple of relevantly identified maximal, submaximal and questionnaire variables. Compared with the results of the rest of hybrid models in this study and the other prediction models in literature, the reported values of R and RMSE have been found to be considerably more accurate. The predictor variables gender, age, maximal heart rate (MX-HR), submaximal ending speed (SM-ES) of the treadmill and Perceived Functional Ability (Q-PFA) questionnaire have been found to be the most relevant variables in predicting VO2max. The results have also been compared with that of Multilayer Perceptron (MLP) and Tree Boost (TB), and it is seen that SVM significantly outperforms other regression methods for prediction of VO2max.

Machine learning and statistical methods for the prediction of maximal oxygen uptake: recent advances.

Maximal oxygen uptake (VO2max) indicates how many milliliters of oxygen the body can consume in a state of intense exercise per minute. VO2max plays an important role in both sport and medical sciences for different purposes, such as indicating the endurance capacity of athletes or serving as a metric in estimating the disease risk of a person. In general, the direct measurement of VO2max provides the most accurate assessment of aerobic power. However, despite a high level of accuracy, practical limitations associated with the direct measurement of VO2max, such as the requirement of expensive and sophisticated laboratory equipment or trained staff, have led to the development of various regression models for predicting VO2max. Consequently, a lot of studies have been conducted in the last years to predict VO2max of various target audiences, ranging from soccer athletes, nonexpert swimmers, cross-country skiers to healthy-fit adults, teenagers, and children. Numerous prediction models have been developed using different sets of predictor variables and a variety of machine learning and statistical methods, including support vector machine, multilayer perceptron, general regression neural network, and multiple linear regression. The purpose of this study is to give a detailed overview about the data-driven modeling studies for the prediction of VO2max conducted in recent years and to compare the performance of various VO2max prediction models reported in related literature in terms of two well-known metrics, namely, multiple correlation coefficient (R) and standard error of estimate. The survey results reveal that with respect to regression methods used to develop prediction models, support vector machine, in general, shows better performance than other methods, whereas multiple linear regression exhibits the worst performance.

Prediction of Maximal Oxygen Uptake From a 3-Minute Walk Based on Gender, Age, and Body Composition

The purpose was to develop new maximal oxygen uptake (VO2max) prediction models using a perceptually regulated 3-minute walk test. VO2max was measured with a maximal incremental cycle test in 283 Japanese adults. A 3-minute walk test was conducted at a self-regulated intensity corresponding to ratings of perceived exertion (RPE) 13. A 3-minute walk distance (3MWD) was significantly related to VO2max (r = .60, P < .001). Three prediction models were developed by multiple regression to estimate VO2max using data on gender, age, 3MWD, and either BMI [BMI model, multiple correlation coefficients (R) = .78, standard error of estimate (SEE) = 5.26 ml·kg-1·min-1], waist circumference (WC model, R = .80, SEE = 5.04 ml·kg-1·min-1), or body fat percentage (%Fat model, R = .84, SEE = 4.57 ml·kg-1·min-1), suggesting that the %Fat model is the best model [VO2max = 37.501 + 0.463 × Gender (0 = women, 1 = men) - 0.195 × Age - 0.589 × %Fat + 0.053 × 3MWD]. Cross-validation by using the predicted residual sum of squares (PRESS) procedures demonstrated a high level of cross-validity of all prediction models. The new VO2max prediction models are reasonably applicable to estimating VO2max in Japanese adults and represent a quick, low-risk, and convenient means for estimating VO2max in the field.

Predicting V˙O2max with an Objectively Measured Physical Activity in Japanese Women

To investigate the use of the accelerometer-determined physical activity (PA) intensity variables as the objective PA variables for estimating VO2max in Japanese adult women. The subjects of this study were 148 Japanese women aged 20 to 69 yr. Maximal oxygen uptake (VO2max) was measured with a maximal incremental test on a bicycle ergometer. Daily step counts (SC) and the amount spent in moderate to vigorous PA (MVPA) and vigorous PA (VPA) were measured using accelerometer-based activity monitors for 7 consecutive days. Using data of age, SC, MVPA, or VPA, and either body mass index (BMI) or waist circumference (WC), the nonexercise VO2max prediction models were derived as BMI models(MVPA), WC models(MVPA), BMI models(VPA), and WC models(VPA), and cross-validated by using two separate cross-validation procedures. SC, MVPA, and VPA were significantly related to VO2max (r = 0.43, r = 0.52, and r = 0.58, respectively). The multiple correlation coefficients for the BMI and WC models(MVAP) were 0.83 and 0.85, respectively, and for the BMI and WC models(VPA), they were 0.85 and 0.86, respectively. The SEE was 3.3 and 3.1 mL x kg(-1) x min(-1) for the BMI and WC models(MVPA), respectively, and it was 3.1 and 3.0 mL x kg(-1) x min(-1) for the BMI and WC models(VPA), respectively. All regression models demonstrated a high level of cross-validity supported by the minor shrinkage of the coefficient of determination and the increment of SEE in the predicted residual sum of squares procedure, and by small constant errors for the subgroups of age, SC, and VO2max between 25 and 35 mL x kg(-1) x min(-1). This study demonstrated that multiple regression models using data of MVPA or VPA were useful in predicting VO2max for Japanese adult women.

Prediction of JOURNAL/jscr/04.03/00124278-200909000-00017/OV0312_7/v/2021-02-09T093636Z/r/image-png O2max During Cycle Ergometry Based on Submaximal Ventilatory Indicators

There are several equations to predict maximum oxygen consumption (VO2max) from ergometric test variables on different ergometers. However, a similar equation using ventilatory thresholds of ergospirometry in a submaximal test on a cycle ergometer is unavailable. The aim of the present study was to assess the accuracy of VO2max prediction models based on indicators of submaximal effort. Accordingly, 4,640 healthy, nonathlete women ages 20 years and older volunteered to be tested on a cycle ergometer using a maximum incremental protocol. The subjects were randomly assigned to 2 groups: group A (estimation) and group B (validation). From the independent variables of weight in kilograms, the second workload threshold (WT2), and heart rate of the second threshold (HRT2), it was possible to build a multiple linear regression model to predict maximal oxygen consumption (VO2max = 40.302 - 0.497 [Weight] - 0.001 [HRT2] + 0.239 [WT2] in mL O2/kg/min(-1); r = 0.995 and standard error of the estimate [SEE] = 0.68 mL O2/kg/min(-1)). The cross-validation method was used in group B with group A serving as the basis for building the model and the validation dataset. The results showed that, in healthy nonathlete women, it is possible to predict VO2max with a minimum of error (SEE = 1.00%) from submaximal indicators obtained in an incremental test.

Support vector regression and multilayer feed forward neural networks for non-exercise prediction of VO2max

The purpose of this study is to develop non-exercise (N-Ex) VO2max prediction models by using support vector regression (SVR) and multilayer feed forward neural networks (MFFNN). VO2max values of 100 subjects (50 males and 50 females) are measured using a maximal graded exercise test. The variables; gender, age, body mass index (BMI), perceived functional ability (PFA) to walk, jog or run given distances and current physical activity rating (PA-R) are used to build two N-Ex prediction models. Using 10-fold cross validation on the dataset, standard error of estimates (SEE) and multiple correlation coefficients (R) of both models are calculated. The MFFNN-based model yields lower SEE (3.23mlkg−1min−1) whereas the SVR-based model yields higher R (0.93). Compared with the results of the other N-Ex prediction models in literature that are developed using multiple linear regression analysis, the reported values of SEE and R in this study are considerably more accurate. Therefore, the results suggest that SVR-based and MFFNN-based N-Ex prediction models can be valid predictors of VO2max for heterogeneous samples.