The Pulse of Asia 2021 Seoul

Ki‐Chul Sung, Jong‐Won Ha, Hae‐Young Lee

PULSE · 2021-yil

Annotatsiya

Objective The goal of this study is to develop a wearable blood pressure(BP) monitoring system which facilitates non-invasive continuous arterial BP monitoring in daily life.Design and Method The developed system consists of a BP estimation model (S/W) and a chest-worn device (H/W).Clinically applicable BP estimation model was developed using 2.4 million cardiac cycles of two commonly used non-invasive biosignals, electrocardiogram (ECG) and photoplethysmogram (PPG) from 1376 surgical patients (Vital DB, Seoul National University Hospital, Korea).Total 28 waveform features including pulse arrival time(PAT), heart rate, and various PPG morphological parameters, were used to construct models of systolic BP(SBP) and diastolic BP(DBP) using various machine learning algorithms such as linear regression, random forest, artificial neural network (ANN), and recurrent neural network (RNN).A wireless chest-worn device measuring ECG and PPG simultaneously was in-house fabricated and ECG and PPG data from 25 healthy subjects wearing the device in their daily life were acquired.The developed BP estimation models were then applied to this dataset after mapping the chest PPG morphology features to the corresponding finger PPG morphology features through an appropriate transfer function.Results Internal dataset validation of the BP estimation models resulted in mean errors of 0.05 ± 6.92 mmHg for SBP, and 0.05 ± 3.99 mmHg for DBP.External validation of the model using another biosignal database

Maqola ma’lumotlari
MualliflarKi‐Chul Sung, Jong‐Won Ha, Hae‐Young Lee
JurnalPULSE
Nashr sanasi2021-01-01
Jild9
SonSuppl. 1
Betlar1-44
Tilen
DOI10.1159/000517557

PULSE jurnalidan boshqa maqolalar

PULSE — barcha maqolalar