Real-time detection of ventilatory-threshold (VT) and predicting VO2max based on VT

Abstract

In this dissertation, a novel approach to detect ventilatory-threshold (VT) and to predict VO2max is proposed.Conventional methods, which detects VT and predicts VO2max, required laboratory environment with gas respiratory analyzer because of using gas data such as VO2, VCO2. And conventional methods can acquire the results only after completing maximal exercise. In this dissertation, 1) Algorithm for detection of VT in real-time using HR is proposed and 2) Regression model for predicting VO2max based on VT without maximal exercise are proposed.To develop the prediction model and validate the new model, 108 healthy, nontrained subjects(80 male, 28 female), aged 19-58 years, participated in this study. All subjects completed a submaximal treadmill GXT according to modified balke protocol. During exercise testing, VT is detected by V-slope method and VO2max is measured by gas respiratory analyzer. To verify the proposed approach, it was applied to several statistical methods such as paired t-test, Bland-Altman analysis. Multiple regression equation is used to predict VO2max using variables of subject’s physical characteristics and HR through exercise from exercise start to VT. In results of paired t-test, there is no significant difference in both males and females between measured and detected VT (r=0.42, P<0.001; r=0.66, P<0.001, respectively). In case of males, variables such as VO2 variation from exercise start to VT, VO2 at VT level and age were significant in predicting VO2max and new model of predicting VO2max based on VT with R2=0.690, SEE=3.62 was developed. In case of females, variables such as age and HR variation from exercise start to VT were significant in predicting VO2max and new model of predicting VO2max based on VT with R2=0.759, SEE=2.91 was developed.From this dissertation, it can conclude that the proposed approach which monitoring slope of HR during initial 3min was useful in detection VT in real-time and VT was useful factor for healthy adults in predicting VO2max. In this dissertation, proposed approach will be a very promising alternative to detect VT and to predict VO2max in many applications of exercise physiology and prescription for children and older adults who was hard to do maximal exercise. With HR monitoring equipment, it will be possible to detect VT and predict VO2max based on VT in both laboratory and filed environment.