Ballistocardiography : physically-based modeling to bridge physiology and technology

Abstract

The ballistocardiogram (BCG) captures the motion of the center of mass (CoM) of the human body resulting from the blood motion within the circulatory system. The BCG signal reflects the status of the cardiovascular system as a whole and, for this reason, it offers a more holistic evaluation of cardiovascular performance than traditional markers, such as electrocardiography or echocardiography. In addition, the acquisition of BCG signals is not invasive, can be performed with several devices -such as accelerometers, chairs, hydraulic system- and does not require body contact. However, the utilization of the BCG as a clinical diagnosis tool and monitoring method is currently hindered by the absence of standardized methods to link the motion of the CoM of the human body, which constitutes the physiological BCG (pBCG), with the BCG signal acquired with sensing devices, which constitute the measured BCG (mBCG). To address this issue, in the first part of the present work we provide a formal definition of pBCG and mBCG, which will be then utilized to (i) define the physical connection between the mBCG obtained with two sensing devices, i.e. the suspended bed and the load cell system, and the pBCG signal and (ii) reconstruct the individual CoM motion. In the second part of the thesis, we focus on the synergistic combination between the physiology behind the BCG signal and the physics of the sensing devices, which may lead to novel clinical applications. In particular, we propose a cuff-less method for absolute pulse pressure assessment via the synergistic integration of two components, namely (i) theoretical simulations of cardiovascular physiology by means of a mathematical closed-loop model of the cardiovascular system, and (ii) synchronous ECG, SCG and BCG data acquired in our laboratory. Then, we present an evolutionary algorithm aimed at individualizing the closed-loop model of the cardiovascular system, with which we will also provide an estimate of the arterial pressure. Finally, in the last part of the thesis, we draw the conclusion of this study, showing how the integration of the mathematical modeling alongside with clinical studies can improve the understanding of the BCG signal and actively contributing to the development of new clinical monitoring solution.