M.Sc Thesis

M.Sc StudentShaviv Eliya
SubjectModeling and Analysis of the Integrated Work of a
Synchronized Cardiac Assist Device and the Failing
DepartmentDepartment of Autonomous Systems and Robotics
Supervisors ASSOCIATE PROF. Amir Landesberg


Heart Failure (HF) is a severe disease where the cardiac function fails to accommodate the demands of the peripheral organs. Current solutions for the advanced HF population are mechanical devices that replace the function of the heart. These devices may ultimately lead to total dependence on the assist device without significant recovery of the native heart in most of the cases. To simultaneously assist the circulation, and to facilitate the potential recovery of the failing heart, a new assist device, denoted as Synchronized Cardiac Assist Device (SCAD), was developed. This device works in cadence with the dynamics of the failing heart; it ejects blood only during the systolic ejection phase and unloads the heart only during the diastolic phase. The SCAD has to learn and to adapt to changes in the loading conditions, and to the left ventricle ejection profile.

            A novel state-space model of the healthy and failing circulations was developed and integrated with the SCAD. This model is a simulation of a hybrid autonomous system which includes the intracellular control of the sarcomeric contraction, the Windkessel model of the peripheral circulation, the cardiac valves, and the assist device.  An innovative concept model of the physiological auto-regulation of the peripheral resistance was also developed and integrated into the system, as means to better simulate the peripheral compensatory mechanisms and their reaction to the assist device. The model was developed and simulated in Matlab (The MathWorks Inc.).

            This model of the heart, called the truth model, was validated against known phenomena, demonstrating satisfying results of isometric contractions and ejecting beats. Utilization of this model allowed us to study the effects of the SCAD on the cardiovascular function and to predict outcomes of different operational schemes. It was shown that the SCAD improves both systolic and diastolic functions of the failing heart, in conjugation with experimental studies. Interestingly, most of the SCAD volume and work were observed as ventricular unloading, while a smaller portion augmented the systolic ejection. The model of the auto-regulation of peripheral resistance showed the anticipated physiological reaction to the assisted circulation, demonstrated by a decreased afterload.  A unique and discriminating feature of conservation of energy was obtained, without an underlying assumption in model.

             A control system for the SCAD was developed based on the model outputs utilizing the QFT method to compensate the uncertain nonlinear model behavior.  The controllers were designed to modulate the SCAD function to meet the dynamic metabolic demands of the body while adapting to changes in the cardiac contractility. The suggested control system was integrated with the truth model and shown satisfying results for the Single Input Single Output (SISO) case.