Wearable Device for Daily Continuous Blood Pressure Estimation Based on Pulse Rate Measurement

Blood pressure is a most basic vital sign, and recently, the importance of daily continuous blood pressure (DCBP) has been pointed out. In this article we present a new daily DCBP estimation method and its experimental validation and optimization. The novel circulatory system model for DCBP estimation with the input of pulse rate consists of the circulatory dynamics model and the circulatory control inverse model. We present a computational result for 24 h blood flow dynamics with blood pressure, blood flow, and blood volume in left/right atrium/ventricle and pulmonary/systemic arteries/veins. Then we present the experimental validation and optimization of the DCBP estimation method. A 25-h continuous pulse rate measurement using a wearable device and blood pressure measurement with 30 min interval using an ABPM device were simultaneously performed for 29 subjects. Optimum condition of the model parameters determination was obtained in consideration of accuracy and computational load. In the results, mean values \(\pm\) standard deviations of the mean errors for systolic and diastolic pressures were 0.3 \(\pm\) 4.3/-0.1 \(\pm\) 2.7 mmHg, and those of the mean absolute errors were 11.6 \(\pm\) 3.6/8.0 \(\pm\) 2.0 mmHg, respectively. The model parameters can be appropriately determined with about 10 ABPM data. These results show that the present estimation method can be applied to practical blood pressure estimation devices with good accuracy, small computational load, and modest number of calibration data.