Blood Pressure Level and Heart Rate Detection from Photoplethysmography Signals Using DT–CWT

Abstract

In this study, it was aimed to estimate systolic and diastolic blood pressures and heart rate using Photoplethysmography (PPG) signals. The PPG signals data used in the study were obtained from an open database containing signals and information of 219 people. With the help of the Dual Tree Complex Wavelet Transform (DT-CWT) method, The properties such as the average power, absolute value mean, kurtosis, skewness and standard deviation of the coefficients of each frequency subbands were obtained. Regression analysis was performed on the extracted PPG signals using Linear Regression (DR), Random Forest (RF) and Support Vector Machines (SVM) algorithms in the Weka program, and blood pressure levels and heart rates were estimated. As a result of the regression analysis, it was seen that blood pressure and heart rate estimations with a higher correlation coefficient and a lower average margin of error, heart rate and diastolic blood pressure analysis with the RF algorithm using the DT-CWT method, and systolic blood pressure analysis with the SVM algorithm would be more accurate.