Separation of Long-term and Short-term Variabilities in Blood Pressure Estimation from Photoplethysmograph

Abstract

In this paper, we propose an extended method to apply the longtime blood-pressure (BP) monitoring in the BP variability estimation without a cuff. Our proposed method separates into long-term and short-term variability by focusing on the temporal factors of BP variation. So, the method consists of previous estimation method, low-pass filtering, and other estimation for short-term variation factor. Although the time series BP measured using a cuff and phoplethysmograph (PPG) are used to create the multiple regression equation, only the features extracted from PPG is used to estimate the BP variability. As an experimental result, the correlation coefficients between grand truth and estimated BP variability was improved from r = 0.35 (previous method) to r = 0.42 (proposed method). In particular, the estimation results for short-term factors showed good accuracy (r = 0.67). These results suggested that our method has feasibility of a simple BP monitoring for a long period.