Continuous Blood Pressure Monitoring Algorithm using Laser Doppler Flowmetry

A novel machine learning algorithm is introduced to estimate continuous blood pressure monitoring using Laser Doppler Flowmetry (LDF). LDF provides instantaneous, continuous, and noninvasive measurements of blood flow in a small tissue sample. The proposed algorithm segments the continuous blood flow profile based on heartbeat cycles to subsequently extract multiple features. The beat-to-beat blood pressure was estimated from a multi-layer neural network algorithm using the extracted features. The algorithm was also validated with clinically proven cuff based continuous blood pressure sensors. Mean average error values of 4.54 \sim5.37 mmHg were observed, which conform to a Grade B/C category per the IEEE standard 1708-2014 for cuffless blood pressure measuring devices.