Limitation of Calibration-free PPG-based Blood Pressure Estimation

Abstract

Photoplethysmography (PPG) is a promising technology for non-invasive and cuffless blood pressure (BP) monitoring. In this context, CSEM has recently explored a feature learning approach to automatically extract relevant information from the PPG waveform and estimate BP accordingly. To date, a calibration procedure remains crucial to reach clinically acceptable accuracy. Although this is a weakness of most PPG-based BP monitoring technologies, this research demonstrates the limitations of a calibration-free model and confirms the importance of a calibration to obtain an accurate estimate of absolute BP.