Cuff-free blood pressure estimation using signal processing techniques

Abstract

Since blood pressure is a significant parameter to examine people's physical attributes and it is useful to indicate cardiovascular diseases, the measurement/estimation of blood pressure has gained increasing attention. The continuous, cuff-less and non-invasive blood pressure estimation is required for the daily health monitoring. In recent years, studies have been focusing on the ways of blood pressure estimation based on other physiological parameters. It is widely accepted that the pulse transit time (PTT) is related to arterial stiffness, and can be used to estimate blood pressure.

A promising signal processing technology, Hilbert-Huang Transform (HHT), is introduced to analyze both ECG and PPG data, which are applied to calculate PTT. The relationship between blood pressure and PTT is illustrated, and the problems of calibration and re-calibration are also discussed. The proposed algorithm is tested based on the continuous data from MIMIC database. To verify the algorithm, the HHT algorithm is compared with other used processing technique (wavelet transform). The accuracy is calculated to validate the method. Furthermore, we collect data using our own developed system and test our algorithm.