Continuous Cuffless Blood Pressure Measurement Using Body Sensors

Non-invasive cuffless blood pressure (BP) measurement using pulse arrival time (PAT) offers some unique features for healthcare applications including improved wearability and user-friendly BP monitoring. Researchers have mostly been using electrocardiography (ECG) and photoplesthysmography (PPG) signals to measure PAT, which requires at least two separate attachments to the subject's body. In this paper, human cuffless systolic BP (SBP) calculations using the shoulders bioimpedance (BImp) signal were investigated and compared to PPG (left ear sensor) and oscillometric sphygmomanometric SBP (left upper arm). The BImp sensors were placed on subject's shoulder and could be completely hidden under their clothes. In order to calculate different BP trends, two groups of experimental conditions (mainly postures and exercises) were followed. Data was obtained from 43 volunteer participants whilst they were sitting, standing and supine, and 26 subjects then cycled on an exercise bike with light, moderate and heavy resistance settings. Four different mathematical models for calculating SBP were compared for accuracy, using both BImp and PPG signal data. Overall, the results of SBP calculation with BImp based PATs were 2% more accurate than that of PATs extracted from PPG.