scholarly journals Temporal convolutional networks predict dynamic oxygen uptake response from wearable sensors across exercise intensities

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Robert Amelard ◽  
Eric T. Hedge ◽  
Richard L. Hughson
Keyword(s):  
Ventilatory Threshold ◽  
Wearable Sensors ◽  
Binary Sequence ◽  
Ground Truth ◽  
Cycle Ergometer ◽  
Convolutional Network ◽  
Metabolic System ◽  
Pseudorandom Binary Sequence ◽  
Temporal Prediction ◽  
Ergometer Exercise

AbstractOxygen consumption ($$\dot{\,{{\mbox{V}}}}{{{\mbox{O}}}}_{2}$$ V ̇ O 2 ) provides established clinical and physiological indicators of cardiorespiratory function and exercise capacity. However, $$\dot{\,{{\mbox{V}}}}{{{\mbox{O}}}}_{2}$$ V ̇ O 2 monitoring is largely limited to specialized laboratory settings, making its widespread monitoring elusive. Here we investigate temporal prediction of $$\dot{\,{{\mbox{V}}}}{{{\mbox{O}}}}_{2}$$ V ̇ O 2 from wearable sensors during cycle ergometer exercise using a temporal convolutional network (TCN). Cardiorespiratory signals were acquired from a smart shirt with integrated textile sensors alongside ground-truth $$\dot{\,{{\mbox{V}}}}{{{\mbox{O}}}}_{2}$$ V ̇ O 2 from a metabolic system on 22 young healthy adults. Participants performed one ramp-incremental and three pseudorandom binary sequence exercise protocols to assess a range of $$\dot{\,{{\mbox{V}}}}{{{\mbox{O}}}}_{2}$$ V ̇ O 2 dynamics. A TCN model was developed using causal convolutions across an effective history length to model the time-dependent nature of $$\dot{\,{{\mbox{V}}}}{{{\mbox{O}}}}_{2}$$ V ̇ O 2 . Optimal history length was determined through minimum validation loss across hyperparameter values. The best performing model encoded 218 s history length (TCN-VO2 A), with 187, 97, and 76 s yielding <3% deviation from the optimal validation loss. TCN-VO2 A showed strong prediction accuracy (mean, 95% CI) across all exercise intensities (−22 ml min−1, [−262, 218]), spanning transitions from low–moderate (−23 ml min−1, [−250, 204]), low–high (14 ml min−1, [−252, 280]), ventilatory threshold–high (−49 ml min−1, [−274, 176]), and maximal (−32 ml min−1, [−261, 197]) exercise. Second-by-second classification of physical activity across 16,090 s of predicted $$\dot{\,{{\mbox{V}}}}{{{\mbox{O}}}}_{2}$$ V ̇ O 2 was able to discern between vigorous, moderate, and light activity with high accuracy (94.1%). This system enables quantitative aerobic activity monitoring in non-laboratory settings, when combined with tidal volume and heart rate reserve calibration, across a range of exercise intensities using wearable sensors for monitoring exercise prescription adherence and personal fitness.

Analysis on Ventilatory Threshold and Muscle Fatigue in an Adolescent with Blindness during Cycle Ergometer Exercise

2009 ◽  
Vol 38 ◽  
pp. 709-718
Author(s):  
Dong Yup Han ◽  
Tae Beom Seo ◽  
Dong Hee Lee ◽  
Kwang Sub Byun ◽  
Il Gue Jeong ◽  
...  
Keyword(s):  
Muscle Fatigue ◽  
Ventilatory Threshold ◽  
Cycle Ergometer ◽  
Cycle Ergometer Exercise ◽  
Ergometer Exercise

Kinetics of Oxygen Uptake for Submaximal Exercise in Hyperoxia, Normoxia, and Hypoxia

1995 ◽  
Vol 20 (2) ◽  
pp. 198-210 ◽  
Author(s):  
Richard L. Hughson ◽  
John M. Kowalchuk
Keyword(s):  
Ventilatory Threshold ◽  
Binary Sequence ◽  
Submaximal Exercise ◽  
Work Rate ◽  
Oxygen Deficit ◽  
Gas Mixture ◽  
Pseudorandom Binary Sequence ◽  
Upper Limit ◽  
Kinetics Of ◽  
Hypoxic Gas Mixture

This study evaluated the dynamic response of [Formula: see text] in 6 healthy men at the onset and end of submaximal step changes in work rate during a pseudorandom binary sequence (PRBS) exercise test and during ramp incremental exercise to exhaustion while breathing three different gas mixtures. The fractional concentrations of inspired O2 were 0.14, 0.21, and 0.70 for the hypoxic, normoxic, and hyperoxic tests, respectively. Both maximal [Formula: see text] and work rate was significantly reduced in hypoxic tests compared to normoxic and hyperoxic tests. Maximal work rate was greater in hyperoxia than in normoxia. Work rate at ventilatory threshold was lower in hypoxia than in normoxia and hyperoxia but above the upper limit of exercise for the submaximal tests. Hypoxia significantly slowed the response of [Formula: see text] both at the onset and end of exercise compared to normoxia and hyperoxia. Hypoxia also modified the response to PRBS exercise, and again there was no difference between normoxia and hyperoxia. These data support the concept that [Formula: see text] kinetics can be slowed from the normoxic response by a hypoxic gas mixture. Key words: [Formula: see text]max, ventilatory threshold, oxygen deficit, pseudorandom binary sequence

scholarly journals Effects of music, video, and 360‐degree video on cycle ergometer exercise at the ventilatory threshold

2019 ◽  
Vol 29 (8) ◽  
pp. 1161-1173 ◽  
Author(s):  
Jonathan M. Bird ◽  
Costas I. Karageorghis ◽  
Steven J. Baker ◽  
David A. Brookes
Keyword(s):  
Ventilatory Threshold ◽  
Cycle Ergometer ◽  
Music Video ◽  
Cycle Ergometer Exercise ◽  
Ergometer Exercise

O2 uptake kinetics after acetazolamide administration during moderate- and heavy-intensity exercise

1998 ◽  
Vol 85 (4) ◽  
pp. 1384-1393 ◽  
Author(s):  
Barry W. Scheuermann ◽  
John M. Kowalchuk ◽  
Donald H. Paterson ◽  
David A. Cunningham
Keyword(s):  
Energy Demand ◽  
Ventilatory Threshold ◽  
Venous Blood ◽  
Lactate Concentration ◽  
Exponential Model ◽  
Constant Load ◽  
Cycle Ergometer ◽  
Work Rate ◽  
Ergometer Exercise ◽  
Single Response

Inhibition of carbonic anhydrase (CA) is associated with a lower plasma lactate concentration ([La−]pl) during fatiguing exercise. We hypothesized that a lower [La−]plmay be associated with faster O2uptake (V˙o 2) kinetics during constant-load exercise. Seven men performed cycle ergometer exercise during control (Con) and acute CA inhibition with acetazolamide (Acz, 10 mg/kg body wt iv). On 6 separate days, each subject performed 6-min step transitions in work rate from 0 to 100 W (below ventilatory threshold, <V˙eT) or to a V˙o 2 corresponding to ∼50% of the difference between the work rate atV˙eT and peakV˙o 2(>V˙eT). Gas exchange was measured breath by breath. Trials were interpolated at 1-s intervals and ensemble averaged to yield a single response. The mean response time (MRT, i.e., time to 63% of total exponential increase) for on- and off-transients was determined using a two- (<V˙eT) or a three-component exponential model (>V˙eT). Arterialized venous blood was sampled from a dorsal hand vein and analyzed for [La−]pl. MRT was similar during Con (31.2 ± 2.6 and 32.7 ± 1.2 s for on and off, respectively) and Acz (30.9 ± 3.0 and 31.4 ± 1.5 s for on and off, respectively) for work rates <V˙eT. At work rates >V˙eT, MRT was similar between Con (69.1 ± 6.1 and 50.4 ± 3.5 s for on and off, respectively) and Acz (69.7 ± 5.9 and 53.8 ± 3.8 s for on and off, respectively). On- and off-MRTs were slower for >V˙eT than for <V˙eT exercise. [La−]plincreased above 0-W cycling values during <V˙eT and >V˙eT exercise but was lower at the end of the transition during Acz (1.4 ± 0.2 and 7.1 ± 0.5 mmol/l for <V˙eT and >V˙eT, respectively) than during Con (2.0 ± 0.2 and 9.8 ± 0.9 mmol/l for <V˙eT and >V˙eT, respectively). CA inhibition does not affect O2 utilization at the onset of <V˙eT or >V˙eT exercise, suggesting that the contribution of oxidative phosphorylation to the energy demand is not affected by acute CA inhibition with Acz.

Alteration of plasma endothelin-1 by exercise at intensities lower and higher than ventilatory threshold

1994 ◽  
Vol 77 (3) ◽  
pp. 1399-1402 ◽  
Author(s):  
S. Maeda ◽  
T. Miyauchi ◽  
K. Goto ◽  
M. Matsuda
Keyword(s):  
Plasma Concentration ◽  
Ventilatory Threshold ◽  
Vascular Endothelial Cells ◽  
Plasma Concentrations ◽  
Cycle Ergometer ◽  
Intercollegiate Athletes ◽  
Endothelin 1 ◽  
Ergometer Exercise ◽  
Before And After ◽  
Potent Vasoconstrictor

The purpose of this study was to investigate whether the release of endothelin-1 (ET-1), a potent vasoconstrictor peptide produced by vascular endothelial cells, is induced by exercise. Venous plasma concentrations of ET-1 were measured by sandwich-enzyme immunoassay before and after endurance exercise with a cycle ergometer at different intensities. Male intercollegiate athletes participated in the study and performed cycle ergometer exercise of 30 min duration at intensities of 90 or 130% of their individual ventilatory threshold (VT). The plasma concentration of ET-1 was slightly but significantly increased after exercise at 90% and markedly increased after exercise at 130% of individual VT. The increase in ET-1 was greatest 30 min after exercise at both intensities. It was first demonstrated that the plasma concentration of ET-1 was significantly increased after exercise: the greater the intensity, the greater the extent of the increase in plasma ET-1 concentration. However, the precise physiological roles of ET-1 during exercise remain to be elucidated.

Association between HMOX-1 genotype and cardiac function during exercise

2008 ◽  
Vol 33 (3) ◽  
pp. 450-460 ◽  
Author(s):  
Zihong He ◽  
Yang Hu ◽  
Lianshi Feng ◽  
Yingli Lu ◽  
Dapeng Bao ◽  
...  
Keyword(s):  
Ventilatory Threshold ◽  
Preliminary Evidence ◽  
Cycle Ergometer ◽  
Cardiac Response ◽  
Heme Oxygenase 1 ◽  
Young Males ◽  
Genotype Effect ◽  
Ergometer Exercise ◽  
Study Population ◽  
Training State

The human gene for heme oxygenase-1 (HMOX-1) plays an important role in the regulation of cardiovascular function and its adaptive response to a variety of stressors. The purpose of this study was to examine the possible association between HMOX-1 genotypes (for –1135A/G, –413A/T, and rs5755720 polymorphisms) and cardiac structural and functional parameters at rest and during submaximal cycle-ergometer exercise (50, 100, and 150 W) in a pre-training state (baseline) and after endurance training (18 weeks, 95%~105% individual ventilatory threshold). The study population consisted of 102 Chinese young males (non-athletes) of Han origin. For the –1135A/G polymorphism, we found a significant genotype effect (p < 0.05) in cardiac output (Q) corrected for body surface area (BSA; Q·BSA–1) at 50 W and stroke volume (SV) corrected for BSA (SV·BSA–1) at 100 W. For the –413A/T polymorphism, we found a significant genotype effect (p < 0.05) in ejection fraction (EF) at 100 W. For the rs5755720 polymorphism, we found a significant genotype effect (p < 0.01 or p < 0.05) in most variables (Q·BSA–1 across all workloads, SV·BSA–1 at 100 W, and EF at 50 and 100 W). Briefly, rs5755720 individuals with a CC genotype presented overall higher values in the different cardiac variables than their CT and (or) TT counterparts. In summary, although more research is needed with diseased populations and other ethnic groups, we found preliminary evidence of an association between cardiac response to submaximal exercise and HMOX-1 genotype. The present preliminary findings could provide insights to future studies searching for cardioprotective genotypes.

Delayed vagal withdrawal slows circulatory but not oxygen uptake responses at work increase

1998 ◽  
Vol 274 (5) ◽  
pp. R1268-R1273 ◽  
Author(s):  
Naoyuki Hayashi ◽  
Ayumu Tanaka ◽  
Mutsuhisa Ishihara ◽  
Takayoshi Yoshida
Keyword(s):  
Oxygen Uptake ◽  
Cold Water ◽  
Ventilatory Threshold ◽  
Muscle Blood Flow ◽  
Cycle Ergometer ◽  
Vagal Activity ◽  
Mean Response Time ◽  
Sympathetic Vasoconstriction ◽  
Ergometer Exercise ◽  
The Face

The effect of delayed vagal activity withdrawal on cardiorespiratory responses at an increase in workload was examined. Eleven volunteers (21 ± 3 yr, 66 ± 4 kg) performed cycle ergometer exercise at a work rate corresponding to 80% of ventilatory threshold after 3 min of unloaded cycling. Facial stimulation was given by applying a vinyl bag filled with cold water (3–5°C) to the face 1 min before to 1 min after the increase in workload (S2 trial) or no stimulation was given (Nr trial). Oxygen uptake (V˙o 2), heart rate (HR), and cardiac output (Q˙) were continuously recorded in four transitions for each trial. Data were averaged for each subject and trial. Mean response time (MRT, sum of delay and time constant) was calculated with a monoexponential fitting. Facial stimulation induced acute bradycardia (−10 ± 5 beats/min in S2 trial). The MRT of HR and Q˙ was significantly longer in the S2 trials (46 ± 35 and 37 ± 23 s) than in the Nr trials (26 ± 18 and 28 ± 19 s, respectively), but no significant change inV˙o 2 MRT was shown (36 ± 7 vs. 38 ± 12 s). These findings suggest that increased vagal activity delays the central circulatory responses, which does not alter the V˙o 2 kinetics at the onset of stepwise increase in workload. The maintenance ofV˙o 2 kinetics during acute bradycardia may either reflect the fact that some intramuscular processes (such as oxidative enzyme inertia) limitV˙o 2 kinetics or alternatively that increased sympathetic vasoconstriction at some remote site defends exercising muscle blood flow.

Investigation of VO2 kinetics in humans with pseudorandom binary sequence work rate change

1990 ◽  
Vol 68 (2) ◽  
pp. 796-801 ◽  
Author(s):  
R. L. Hughson ◽  
D. A. Winter ◽  
A. E. Patla ◽  
G. D. Swanson ◽  
L. A. Cuervo
Keyword(s):  
Fourier Analysis ◽  
Mean Square Error ◽  
Binary Sequence ◽  
Frequency Component ◽  
Cycle Ergometer ◽  
Work Rate ◽  
High Frequency Component ◽  
Mean Square ◽  
Pseudorandom Binary Sequence ◽  
The Mean

The dynamic response of oxygen uptake (VO2) was investigated with two different cycle ergometer tests in which the work rate changed as a pseudorandom binary sequence (PRBS). One sequence had 15 units, each of 30-s duration for a total of 450 s (PRBS1). The second had 63 units, each of 5-s duration for a total of 315 s (PRBS2). The useful range of frequencies available for investigation of the dynamic characteristics of the VO2 response as described by their bandwidth were 0.002-0.013 Hz for PRBS1 and 0.003-0.089 Hz for PRBS2. Eight subjects each completed both PRBS tests. Data from four or five consecutive sequences were ensemble averaged to reduce the biological noise. A Fourier analysis was then conducted, with the range of frequencies investigated spanning those of the bandwidth for PRBS2. This was up to the 28th harmonic. For PRBS1, the VO2 response could be adequately reconstructed by including Fourier coefficients only up to the 5th harmonic. In contrast, for PRBS2, there was still a clear pattern in the residuals at the 5th harmonic. The data were not adequately reconstructed until higher-frequency components up to the 28th harmonic were included. Evidence for this came from analysis of the mean square error. The mean square error at the 28th harmonic was reduced to 83 +/- 8% of the mean square error at the 5th harmonic for PRBS1 and to 31 +/- 3% for PRBS2 (P less than 0.0001). These data obtained by Fourier analysis and reconstructed for comparison with the original VO2 response indicate the presence of a high-frequency component that was not apparent when a test with a smaller bandwidth was used as the work rate forcing.

Sign in / Sign up

Export Citation Format

Share Document