Frequency domain analysis to extract dynamic response characteristics for oxygen uptake during transitions to moderate- and heavy-intensity exercises

Longitudinal assessment of cardiorespiratory fitness and body mass of young healthy adults during COVID-19 pandemic

Physical activity was reduced during the COVID-19 pandemic, especially when lockdowns were mandated; however, little is known about the impact of these lifestyle changes on objective measures of cardiorespiratory fitness. To address this knowledge gap, we evaluated the cardiorespiratory fitness of 14 young healthy adults (4 women, age: 27±6 yr) just prior to the pandemic and after ~1 year of public health measures being in place. During fitness assessments, participants performed submaximal pseudorandom cycling exercise to assess cardiorespiratory kinetics, and a 25 W·min^-1 ramp-incremental cycling test to determine peak oxygen uptake (V̇O₂). Cluster analysis identified two subgroups of participants: those who had reduced peak V̇O₂ at the 1-year follow-up (-0.50±0.17 L·min^-1) and those whose peak V̇O₂ was maintained (0.00±0.10 L·min^-1). Participants with reduced peak V̇O₂ also exhibited slower heart rate kinetics (interaction: p=0.01), reduced peak O₂ pulse (interaction: p<0.001), and lower peak work rate (interaction: p<0.001) after 1 year of the pandemic, whereas these variables were unchanged in the group of participants who maintained peak V̇O₂. Regardless of changes in peak V̇O₂, both subgroups of participants gained body mass (main effect: p=0.002), which was negatively correlated with participants' level of self-reported physical activity level at the follow-up assessment (mass: ρ=-0.59, p=0.03) These findings suggest that some young healthy individuals lost cardiorespiratory fitness during the pandemic, while others gained weight, but both changes could potentially increase the risk of adverse health outcomes and disease later in life if left unaddressed.

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

Oxygen consumption ([Formula: see text]) provides established clinical and physiological indicators of cardiorespiratory function and exercise capacity. However, [Formula: see text] monitoring is largely limited to specialized laboratory settings, making its widespread monitoring elusive. Here we investigate temporal prediction of [Formula: see text] 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 [Formula: see text] 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 [Formula: see text] dynamics. A TCN model was developed using causal convolutions across an effective history length to model the time-dependent nature of [Formula: see text]. 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 [Formula: see text] 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.