Wearable Bio-Impedance Sensing Methods for Continuous Monitoring of Hemodynamic Parameters

Abstract

Continuous monitoring of hemodynamic parameters such as blood pressure (BP) provides significant advantages in predicting future cardiovascular disease. Traditional BP measurement methods are based on a cuff, which is bulky, obtrusive, and not applicable to continuous monitoring. Measurement of blood pulsatile is one of the prominent cuffless methods for continuous BP monitoring. The pulse morphology and pulse transit time (PTT) which is the time taken by the pressure pulse to travel between two points in an arterial vessel are highly correlated with the BP. In this dissertation, we present a new cuffless BP method using an array of wrist-worn bio-impedance (Bio-Z) sensors placed on the wrist arteries to monitor the arterial pressure pulse from the blood volume changes. The Bio-Z sensing method is a non-invasive technique that can measure blood volume changes by injecting alternating current (AC) signal that flows deep into the tissue via a pair of electrodes and then, sensing the potential difference on another pair. We present the design of our custom Bio-Z sensing hardware and electrode array wristband that provide high-quality pulse signals through multi-channel Bio-Z sensing from the wrist. BP is accurately estimated by using the AdaBoost regression model based on selected arterial pressure pulse features. Post-exercise BP was accurately estimated with an average correlation coefficient of 0.77 for the diastolic BP and 0.86 for the systolic BP. In addition, we present a Bio-Z simulation platform that models the tissue and arterial pulse wave using a 3D circuit model based on a time-varying impedance grid. The proposed model will enable researchers to create time-varying blood flow models and rapidly test the effectiveness of the sensing methods and algorithms without the need for extensive experimentation. Furthermore, we propose a new multi-source multi-frequency Bio-Z sensing method that provides more localized pulsatile monitoring for improved PTT. Another Bio-Z method is proposed based on a convolutional neural network (CNN) autoencoder that estimates an accurate arterial pulse signal independent of sensing location from multiple pulse signals. The proposed methods contribute to reliable and accurate continuous monitoring of hemodynamic parameters from wrist-worn devices, which can contribute to more effective monitoring and management of the cardiovascular disease.