Wearable respiratory function detection system based on dynamic humidity sensing: Principle and experimental comparison of lung function detection

Abstract

<sec>Traditional lung function detectors are based on measuring the changes in airflow and pressure during expiration and inspiration to evaluate the respiratory function of the subject. These techniques are mainly based on mechanical differential pressure sensors or turbine sensors which evaluate the lung function of the subject by measuring the ability of the subject to blow and inhale and determine their lung function parameters, including peak expiratory flow (PEF) and forced vital capacity (FVC). In this study, we present a wearable respiratory function testing system called the wearable respiratory spectrometer, which is developed based on dynamic humidity sensing technology. By exploring the principles and quantitative design of respiratory detection and conducting simulations of humidity sensors, we investigate the comprehensive characteristics of the system. According to Darcy’s law, the gas flow measured by the wearable respiratory spectrometer is directly proportional to the pressure difference inside and outside the device, showing that the system follows the differential pressure sensing principle. According to this basis and combining the structural characteristics of the system, we establish a quantitative relationship among PEF, FVC, and the changes in sensor electrical signals.</sec><sec>The experimental results validate a linear positive correlation between the maximum rate of relative humidity change inside the spectrometer and PEF. Additionally, the results of simulated moisture volume experiments of the spectrometer show that in the measurement range from 180 to 840 L/min, the indication error of PEF is less than 10%, the adjacent test error is less than 5%, and the frequency response test error is less than 12%, which meet the industry standards for peak expiratory flow meters. Moreover, we compare the spectrometer with traditional portable lung function testing devices in simulated moisture volume experiments at different PEFs (300 to 720 L/min) and FVCs (3 to 6 L) . The results demonstrate that the average indication error of measured PEF and FVC by the spectrometer are about 0.35% and 0.23%, respectively, both are much lower than those of the portable lung function testing devices, thus fully verifying the accuracy and reliability of this system for real-time lung function assessment. Importantly, under simulated free-breathing conditions (PEF from 12 to 24 L/min, FVC from 0.5 to 0.7 L), the changes in the electrical signals of the spectrometer maintain a linear relationship with the moisture volume. Therefore, the wearable respiratory spectrometer can provide the long-term, free, dynamic, and quantitative monitoring of natural and weak nasal breathing. The measured respiratory spectra of subjects have great potential in real-time monitoring of lung function and remote monitoring of respiratory system diseases.</sec>