Blood pressure prediction method using PPG sensors

摘要

The present invention relates to a method of predicting blood pressure (BP) from the photoplethysmogram (PPG) signal associated with a new original data set. The dataset is based on the fact that subjects are periodically monitored for weeks, while public datasets consider short acquisition periods. The predictive method uses key frequencies in the spectrum of the isolated PPG signal using the LASSO algorithm. The effectiveness of the method was evaluated on experimental data recorded over a long period. In addition, an evaluation of the different time markers of the PPG signal that have been proposed in the literature was carried out on the same dataset. These evaluations have shown that only a few of these time markers are useful for the prediction of systolic and diastolic blood pressure. The results highlight that better blood pressure predictions are obtained by using the spectrum of the PPG signal rather than optimally selected time markers.