Estimation of Instantaneous Oxygen Uptake During Exercise and Daily Activities Using a Wearable Cardio-Electromechanical and Environmental Sensor

• Published in: IEEE journal of biomedical and health informatics Vol. 25; no. 3; pp. 634 - 646
• Main Authors: Shandhi, Md Mobashir Hasan, Bartlett, William H., Heller, James Alex, Etemadi, Mozziyar, Young, Aaron, Plotz, Thomas, Inan, Omer T.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.03.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Atmospheric measurements; Atmospheric models; Biomedical monitoring; COSMED K5; EKG; Electrocardiography; Energy expenditure; Estimation; Exercise; Feature extraction; Fitness equipment; Heart rate; Humans; Learning algorithms; Legged locomotion; Low cost; Machine learning; metabolic; Monitoring; Oxygen; Oxygen Consumption; Oxygen uptake; Protocol; Protocols; Seismocardiography; Signal processing; Sternum; Treadmills; Walking; Wearable Electronic Devices; Wearable technology
• ISSN: 2168-2194; 2168-2208
• DOI: 10.1109/JBHI.2020.3009903

Objective: To estimate instantaneous oxygen uptake <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula> 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 <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula> data obtained from the COSMED system. Results: In estimating instantaneous <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula>, we achieved our best results on the treadmill protocol using a combination of SCG (frequency) and AP features (RMSE of 3.68 <inline-formula><tex-math notation="LaTeX">\pm</tex-math></inline-formula> 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 <inline-formula><tex-math notation="LaTeX">\pm</tex-math></inline-formula> 1.47 ml/kg/min and R 2 of 0.64). In estimating <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula> consumed over one minute intervals during the protocols, our median percentage error was 15.8<inline-formula><tex-math notation="LaTeX">\text{}\%</tex-math></inline-formula> for the treadmill protocol and 20.5<inline-formula><tex-math notation="LaTeX">\text{}\%</tex-math></inline-formula> for the outside protocol. Conclusion: SCG, ECG and AP signals from a small wearable patch can enable accurate estimation of instantaneous <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula> 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 <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula>. Significance: Accurate estimation of <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula> with a low cost, minimally obtrusive wearable patch can enable the monitoring of <inline-formula><tex-math notation="LaTeX">{\rm VO}_{2}</tex-math></inline-formula> and EE in everyday settings and make the many applications of these measurements more accessible to the general public.