Measuring Cardiorespiratory Information in Sitting Position using Multiple Piezoelectric Sensors

Abstract

We have been studying equipment to easily acquire cardiorespiratory information at home using piezoelectric sensors arranged on the seat surface of a chair. In our previous study, we suggested that the cardiac and respiratory components could be extracted by executing template matching using a two-dimensional cross-correlation function for the signals that were obtained from the piezoelectric sensors. However, there was a difficulty with the signal extraction, depending on the seating position. Therefore, in this study, we examined the measurement of the heartbeat and breathing interval using independent component analysis and multiple piezoelectric sensors. Moreover, the heartbeat and breathing intervals that were obtained from the extracted cardiorespiratory components using our developed automatic decision method were compared with those obtained from electrocardiogram and pneumogram. As a result, it was found that we could achieve better error rates (0.93±0.44% and 5.23±3.04% for the heartbeat and respiratory intervals, respectively) than in our previous study.