Enhanced implantable device evaluation using a hardware-in-the-loop circulatory simulator

As the number of patients experiencing severe heart failure increases, so too does the need and use of implantable mechanical circulatory support devices, such as ventricular assist devices (VADs) and total artificial hearts (TAHs), for patient treatment. These devices assist the failing heart by taking over the blood pumping process either partially or completely and must be properly vetted and verified for human use by undergoing many years of testing, both using in vitro benchtop and in vivo animal testing. To expedite the laboratory testing process, hardware-in-the-loop (HIL) systems called mock circulation loops (MCLs) have been developed that can test assist devices within a mechanical realization of the human cardiovascular environment. This research focuses on the evaluation of nominal and enhanced VAD function using a hybrid mock circulation loop (hMCL) as a HIL test platform. The research can be split into three major aims: (1) hMCL construction and performance characterization for nominal VAD testing, (2) hMCL based evaluation of enhanced VAD onboard estimation algorithms, (3) Method for HIL implementation of arrhythmic and VAD induced cardiac events in hMCL. Regarding Aim 1, test results of the current hMCL design show that root-mean-square error between simulated and realized physiological pressures across a range of VAD flowrates can be maintained within 1.5-3.5 [mmHg]. The hMCL is able to simulate different levels of patient cardiovascular health and basic sensitivity tests indicate responsiveness to changes in the simulated model parameters; a necessary requirement to accomplish Aim 2. Completed work towards Aim 2 involves evaluating the onboard sensing algorithms being developed on VADs using sensor-based estimation of systemic vascular resistance (SVR) as an example. Preliminary experiments using the hMCL showed that SVR value estimates were accurate within 1.3% and 0.7% compared to the set numerical model values for tests run on continuous and pulsatile flow VADs, respectively. Lastly, methods for incorporating arrhythmic cardiac events and valvular stenosis have been presented towards Aim 3. Preliminary results show less than 2% and 4% mean percent error between aortic and left ventricular pressure tracking, respectively, as well as good agreement between referenced and measured frequency content.