Vital Signs Monitoring System Based on EMFi Sensors and Wavelet Analysis

The paper presents a monitoring system for non-obtrusive measuring of vital signs such as respiration and heart activity. Two sensing channels including electromechanical film sensors (EMFi sensor) tapped on an office chair are used to obtain the ballistocardiographyc (BCG) signals from human subjects. The conditioning circuits include amplification stages (charge amplifier, programmable gain amplifier) and 20 Hz low-pass and 50 Hz notch active filters used to improve the signal-to-noise ratio (SNR). The gain control and analog-to-digital conversion is performed using a multifunction I/O device USB compatible connected to a laptop PC. The digital processing of the acquired signals is an important part of the work. It includes digital filtering, based on Stationary Wavelet Transform, and signal decomposition, based on Discrete Wavelet Transform, in order to extract the respiration and heart rate based on BCG signal analyze. A practical approach concerning the dependence of the used wavelet type on vital signs estimation accuracy is reported. The results on respiration rate and heart rate using BCG signal and the implemented algorithms are compared with the values of these parameters measured through ECG and spirometry analysis.