Study of perfusion based theoretical model and experimental evaluation for wearable CBT measurement

Abstract

Continuous and non-invasive core body temperature (CBT) monitoring is imperative to clinical diagnosis and health monitoring. Current wearable non-invasive CBT measurements ignore the role of human thermoregulation and compromise measurement accuracy under changing environmental conditions that often occur in daily wearable scenarios. In this study, a novel wearable measurement method with optimized parameters (only 1 calibration parameter) was proposed, which could monitor CBT precisely under changing environmental conditions, by improving the sensor structure and quantifying the effect of human thermoregulation directly. Firstly, the proposed method was simulated and analyzed based on finite element method (FEM), and the mean absolute error (0.095 °C) of the model over a wide range of ambient temperatures was reduced by 98.19 % compared to the traditional model. Then, two experiments under ordinary life scenarios were designed to evaluate the method: stable and changing environment experiments had the mean absolute error of 0.04 ± 0.05 °C and 0.19 ± 0.24 °C respectively. Overall, the proposed method can suppress the estimation errors caused by human thermoregulation and has the potential to continuously monitor CBT under changing environmental conditions in daily life.