A Radar Contactless Blood Pressure Detection Technology based on Pulse Waves Using High-order Polynomial Regression

Compared with cuff-contact Blood Pressure (BP) measurement methods such as mercury sphygmomanometers, radar contactless BP detection technology has good application prospects in fields for patients with burns, infectious diseases, or homecare health monitoring due to its ability to continuously monitor BP for a long time without any constraints. Currently, research in this area is mainly based on the pulse transit time (PTT). In this paper, a PTT BP fitting method based on a high-order polynomial regression model was proposed, which uses a 24-GHz continuous wave radar to collect 42 sets of radial artery data from two subjects for seven days. Three regression models were applied for BP calculation and the results were shown. Compared with linear regression and logarithmic regression, the proposed method performed better in terms of the R-square, Root Mean Square Error(RMSE), and Mean Absolute Error(MAE). The Mean Deviation(MD) and Standard Deviation(SD) of systolic and diastolic BP were -0.106 ± 0.474 mmHg and 0.574 ± 2.161 mmHg, respectively. The feasibility and accuracy of this method was demonstrated.