Physiological control of an implantable rotary blood pump / Mahdi Mansouri

• Main Author: Mahdi , Mansouri
• Format: Thesis
• Published: 2016
• Subjects: Q Science (General); T Technology (General)
• Online Access: http://studentsrepo.um.edu.my/6328/1/mahdi.pdf; http://studentsrepo.um.edu.my/6328/
• Institution: Universiti Malaya

Left ventricular assist devices (LVADs) are mechanical pumps that their usage expanded from bridging to recovery to bridging to decision, and destination therapy. A physiologically responsive pump control strategy, which automatically adjusts variations in the metabolic demands, is tremendously needed to maximize the quality of the implant recipients’ life. The aim of this dissertation is providing a robust physiological based controller, which could resist against all possible distortion during its working life. At the first step, the performance of a number of previously proposed physiologically responsive controllers were comparably evaluated. The study proved applying a constant (static) controlling method could not provide the pump best controlling performance level, and indicates the demand of an adaptive (dynamic) controller, which will satisfy all physiological requisites. The results also suggested putting the focus on preload sensitivity of the ventricular myocardium. Such issue is an essential requirement for the Frank-Starling mechanism by which the left ventricular end-diastolic pressure (PLVED) controls the force of contraction of the left ventricle (LV) in proportion to the blood flow received from the right heart and pulmonary circulation. At the next step a preload-based Starling-like controller for Implantable rotary blood pumps (IRBPs) using PLVED as the feedback variable was evaluated in a validated numerical model. The controller emulated the response of the natural LV to changes in PLVED. It was reported the performance of the preload-based Starling-like controller in comparison with recently designed pulsatility controller and constant speed operation. In handling the transition from a baseline state to test states, which included vigorous exercise, blood loss and a major reduction in the LV contractility (LVC), the preload controller outperformed pulsatility control and constant speed operation in all three test scenarios. iv The study third objection was realizing preload-based using a controlling technique that reinforced the system to rapidly reach the target pump flow within eight heartbeats, else suction might occur. The technique must be also robust against noises contaminated the feedback signal. Accordingly, this study also attentively examined the transient and steady state response of two different preload-based control implementations in the numerical model. The implementations were tested under both ideal (noise free) and noisy conditions at baseline to vigorous exercise as well as blood loss transitions. Proportional-integral-derivative (PID) and sliding mode controller (SMC) were the chosen controlling techniques, selected due to their popularity and robustness reputation. While at the noise free condition system measured PLVED was directly fed to the controller, the author contaminated the feedback signal with different levels of Gaussian white noises to realize the noisy condition. The results showed no significant difference between the two preload-based control implementations under ideal condition during all testing scenarios. Proceeding the tests showed that both PID and SMC delivered almost comparable performance at signal to noise ratio (SNR) of 15dB, although by increasing the noise level to 7dB PID finally failed at blood loss scenario and severely penetrated in the suction, indicated by persistent negative PLVED. On the contrary, SMC is the control strategy that not only could tolerate all the noise levels and never fell into the suction region, but also could maintain a reasonable level of hemodynamic parameters deviations comparably. The last objective of this study was to develop an in-vitro evaluation protocol for control system utilizing a mock circulation loop (MCL) exploiting the same scenarios used in the second objective. The test showed that the devised scenarios were useful for evaluation of preload-based control. Observing the results showed the preload controller could again outperform the constant speed operational in all three scenarios, provided the impetus for further animal trials.