Integration of Wearable Sensors and Metabolic Chamber Data for Energy Expenditure Assessment: A Preliminary Study

Accurate assessment of energy expenditure is crucial for characterizing human metabolism and health. The gold standard method for measuring energy expenditure is indirect calorimetry in a metabolic chamber, which provides accurate estimates of metabolic rate under controlled conditions. However, the limited availability of metabolic chambers worldwide limits large-scale research and clinical applications. Wearable technologies offer a promising alternative, allowing continuous monitoring of key physiological parameters such as heart rate variability, electrodermal activity, and physical activity, all of which correlate with metabolic rate. This study integrates 24-hour metabolic chamber measurements and multimodal wearable sensor data to quantify day-night (diurnal-nocturnal) differences in energy expenditure and autonomic state. Physiological data were collected using two multimodal wearable devices: a textile-based smart t-shirt (Wearable Wellness System, Smartex) and a wrist-worn smartwatch (Embrace Plus, Empatica). In a preliminary sample of six healthy men, the median metabolic rate was 42% higher during the daytime compared to nighttime, mirroring a 3.1-fold increase in sympathetic electrodermal activity power during the day. During daytime, metabolic rate showed the strongest pooled Spearman correlations with heart rate variability features derived from the smart t-shirt: mean RR interval (average time between heartbeats; <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\rho=-0.37$</tex>), low-frequency <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\rho=-0.31)$</tex>, and high-frequency <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\rho=-0.28)$</tex> components. At night, tonic electrodermal activity (EDA) features from the smartwatch (Embrace Plus) were most informative, with the minimum (minTonic, lowest baseline skin conductance, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\rho=0.36$</tex>) and mean (meanTonic, average baseline skin conductance, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\rho=0.34$</tex>) values showing the highest associations. Synchronizing wearable-derived measures with continuous chamber calorimetry validates their ability to track diurnal metabolic dynamics and provides a foundation for large-scale metabolic monitoring, facilitating broader applications in health and metabolic research.