학술논문

Estimation of Instantaneous Oxygen Uptake During Exercise and Daily Activities Using a Wearable Cardio-Electromechanical and Environmental Sensor
Document Type
Periodical
Source
IEEE Journal of Biomedical and Health Informatics IEEE J. Biomed. Health Inform. Biomedical and Health Informatics, IEEE Journal of. 25(3):634-646 Mar, 2021
Subject
Bioengineering
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Signal Processing and Analysis
Protocols
Biomedical monitoring
Electrocardiography
Atmospheric measurements
Estimation
Legged locomotion
Heart rate
Seismocardiography
COSMED K5
Oxygen uptake
metabolic
machine learning
Language
ISSN
2168-2194
2168-2208
Abstract
Objective: To estimate instantaneous oxygen uptake ${\rm VO}_{2}$ with a small, low-cost wearable sensor during exercise and daily activities in order to enable monitoring of energy expenditure (EE) in uncontrolled settings. We aim to do so using a combination of seismocardiogram (SCG), electrocardiogram (ECG) and atmospheric pressure (AP) signals obtained from a minimally obtrusive wearable device. Methods: In this study, subjects performed a treadmill protocol in a controlled environment and an outside walking protocol in an uncontrolled environment. During testing, the COSMED K5 metabolic system collected gold standard breath-by-breath (BxB) data and a custom-built wearable patch placed on the mid-sternum collected SCG, ECG and AP signals. We extracted features from these signals to estimate the BxB ${\rm VO}_{2}$ data obtained from the COSMED system. Results: In estimating instantaneous ${\rm VO}_{2}$, we achieved our best results on the treadmill protocol using a combination of SCG (frequency) and AP features (RMSE of 3.68 $\pm$ 0.98 ml/kg/min and R 2 of 0.77). For the outside protocol, we achieved our best results using a combination of SCG (frequency), ECG and AP features (RMSE of 4.3 $\pm$ 1.47 ml/kg/min and R 2 of 0.64). In estimating ${\rm VO}_{2}$ consumed over one minute intervals during the protocols, our median percentage error was 15.8$\text{}\%$ for the treadmill protocol and 20.5$\text{}\%$ for the outside protocol. Conclusion: SCG, ECG and AP signals from a small wearable patch can enable accurate estimation of instantaneous ${\rm VO}_{2}$ in both controlled and uncontrolled settings. SCG signals capturing variation in cardio-mechanical processes, AP signals, and state of the art machine learning models contribute significantly to the accurate estimation of instantaneous ${\rm VO}_{2}$. Significance: Accurate estimation of ${\rm VO}_{2}$ with a low cost, minimally obtrusive wearable patch can enable the monitoring of ${\rm VO}_{2}$ and EE in everyday settings and make the many applications of these measurements more accessible to the general public.