Non-Invasive Blood Pressure Total Waveform Monitoring Using Information Extracted by an Extended Kalman Filter Algorithm from Pulsations in an Oscillatory Cuff

Abstract

This paper describes a novel approach designed for non-invasive, routine screening of patients for cardiovascular disorders without the complexities of using an electrocardiogram or invasive probes. Specifically, an oscillographic-view of the brachial artery blood pressure waveform, including the dicrotic notch, is extracted from information in the pressure pulsations from an ordinary blood pressure cuff. The novelty of this approach is that the total continuous shape of the waveform, not just two numbers for pressures, is generated. A computer algorithm processes the cuff pressure pulsations and provides a near real-time visual estimate of the continuous shape of the blood pressure waveform to be viewed on oscilloscopes commonly used in hospitals and medical clinics. A model-based Extended Kalman Filter (EKF) algorithm is used to process the information and identify the coefficients of a Fourier transform model for the waveform and the coefficients for an empirical model for artery stiffness. By viewing the total waveform, variations or disorders in the waveform during a series of pulse cycles can be observed and easily recognized. Simulations of two-case studies demonstrate successful convergence of the algorithm for a variety of waveform disorders including arrhythmia, variations in the shape of the dicrotic notch, changing systolic and diastolic pressure levels, and stiff arteries. Experimental verification of this proposed procedure will require invasive pressure measurements while simultaneously processing the algorithm for non-invasive waveform comparisons. Once the algorithm is verified experimentally, the end goal is to use the procedure on patients with known cardiovascular diseases in order to easily create a database of waveform disorders correlated with disorders of specific cardiovascular diseases.