Cuffless and continuous blood pressure estimation based on multiple regression analysis

This paper proposes a new method which utilizes the multiple parameters to estimate blood pressure continuously. Five parameters extracted from ECG and PPG are used to fit the values of systolic blood pressure (SBP), mean arterial pressure (MAP) and diastolic blood pressure (DBP). 10 healthy subjects were recruited to take part in the measurement of ECG, PPG and continuous blood pressure. Three different equations are established to estimate the above three blood pressures by using stepwise regression method. The estimated algorithm has been verified on an independent dataset. The mean error of the estimation is less than 6 mmHg and the standard deviation of the error is less than 6.5 mmHg among all the blood pressure of the dataset. This demonstrates the effectiveness of the proposed method with certain ability to predict the blood pressure. In the near future, we will improve and validate this method on other databases with additional parameters or new statistical methods.