Physiological control of dual rotary pumps as a biventricular assist device using a master/slave approach

An in vitro model to study suction events by a ventricular assist device: validation with clinical data

Introduction: Ventricular assist devices (LVADs) are a valuable therapy for end-stage heart failure patients. However, some adverse events still persist, such as suction that can trigger thrombus formation and cardiac rhythm disorders. The aim of this study is to validate a suction module (SM) as a test bench for LVAD suction detection and speed control algorithms. Methods: The SM consists of a latex tube, mimicking the ventricular apex, connected to a LVAD. The SM was implemented into a hybrid in vitro-in silico cardiovascular simulator. Suction was induced simulating hypovolemia in a profile of a dilated cardiomyopathy and of a restrictive cardiomyopathy for pump speeds ranging between 2,500 and 3,200 rpm. Clinical data collected in 38 LVAD patients were used for the validation. Clinical and simulated LVAD flow waveforms were visually compared. For a more quantitative validation, a binary classifier was used to classify simulated suction and non-suction beats. The obtained classification was then compared to that generated by the simulator to evaluate the specificity and sensitivity of the simulator. Finally, a statistical analysis was run on specific suction features (e.g., minimum impeller speed pulsatility, minimum slope of the estimated flow, and timing of the maximum slope of the estimated flow). Results: The simulator could reproduce most of the pump waveforms observed in vivo. The simulator showed a sensitivity and specificity and of 90.0% and 97.5%, respectively. Simulated suction features were in the interquartile range of clinical ones. Conclusions: The SM can be used to investigate suction in different pathophysiological conditions and to support the development of LVAD physiological controllers.

Use of 3D anatomical models in mock circulatory loops for cardiac medical device testing

INTRODUCTION

Mock circulatory loops (MCLs) are mechanical representations of the cardiovascular system largely used to test the hemodynamic performance of cardiovascular medical devices (MD). Thanks to 3 dimensional (3D) printing technologies, MCLs can nowadays also incorporate anatomical models so to offer enhanced testing capabilities. The aim of this review is to provide an overview on MCLs and to discuss the recent developments of 3D anatomical models for cardiovascular MD testing.

METHODS

The review first analyses the different techniques to develop 3D anatomical models, in both rigid and compliant materials. In the second section, the state of the art of MCLs with 3D models is discussed, along with the testing of different MDs: implantable blood pumps, heart valves, and imaging techniques. For each class of MD, the MCL is analyzed in terms of: the cardiovascular model embedded, the 3D model implemented (the anatomy represented, the material used, and the activation method), and the testing applications.

DISCUSSIONS AND CONCLUSIONS

MCLs serve the purpose of testing cardiovascular MDs in different (patho-)physiological scenarios. The addition of 3D anatomical models enables more realistic connections of the MD with the implantation site and enhances the testing capabilities of the MCL. Current attempts focus on the development of personalized MCLs to test MDs in patient-specific hemodynamic and anatomical scenarios. The main limitation of MCLs is the impossibility to assess the impact of a MD in the long-term and at a biological level, for which animal experiments are still needed.

Mock circulatory loop applications for testing cardiovascular assist devices and in vitro studies

The mock circulatory loop (MCL) is an in vitro experimental system that can provide continuous pulsatile flows and simulate different physiological or pathological parameters of the human circulation system. It is of great significance for testing cardiovascular assist device (CAD), which is a type of clinical instrument used to treat cardiovascular disease and alleviate the dilemma of insufficient donor hearts. The MCL installed with different types of CADs can simulate specific conditions of clinical surgery for evaluating the effectiveness and reliability of those CADs under the repeated performance tests and reliability tests. Also, patient-specific cardiovascular models can be employed in the circulation of MCL for targeted pathological study associated with hemodynamics. Therefore, The MCL system has various combinations of different functional units according to its richful applications, which are comprehensively reviewed in the current work. Four types of CADs including prosthetic heart valve (PHV), ventricular assist device (VAD), total artificial heart (TAH) and intra-aortic balloon pump (IABP) applied in MCL experiments are documented and compared in detail. Moreover, MCLs with more complicated structures for achieving advanced functions are further introduced, such as MCL for the pediatric application, MCL with anatomical phantoms and MCL synchronizing multiple circulation systems. By reviewing the constructions and functions of available MCLs, the features of MCLs for different applications are summarized, and directions of developing the MCLs are suggested.

Fully Elman Neural Network: A Novel Deep Recurrent Neural Network Optimized by an Improved Harris Hawks Algorithm for Classification of Pulmonary Arterial Wedge Pressure

Heart failure (HF) is one of the most prevalent life-threatening cardiovascular diseases in which 6.5 million people are suffering in the USA and more than 23 million worldwide. Mechanical circulatory support of HF patients can be achieved by implanting a left ventricular assist device (LVAD) into HF patients as a bridge to transplant, recovery or destination therapy and can be controlled by measurement of normal and abnormal pulmonary arterial wedge pressure (PAWP). While there are no commercial long-term implantable pressure sensors to measure PAWP, real-time non-invasive estimation of abnormal and normal PAWP becomes vital. In this work, first an improved Harris Hawks optimizer algorithm called HHO+ is presented and tested on 24 unimodal and multimodal benchmark functions. Second, a novel fully Elman neural network (FENN) is proposed to improve the classification performance. Finally, four novel 18-layer deep learning methods of convolutional neural networks (CNNs) with multi-layer perceptron (CNN-MLP), CNN with Elman neural networks (CNN-ENN), CNN with fully Elman neural networks (CNN-FENN), and CNN with fully Elman neural networks optimized by HHO+ algorithm (CNN-FENN-HHO+) for classification of abnormal and normal PAWP using estimated HVAD pump flow were developed and compared. The estimated pump flow was derived by a non-invasive method embedded into the commercial HVAD controller. The proposed methods are evaluated on an imbalanced clinical dataset using 5-fold cross-validation. The proposed CNN-FENN-HHO+ method outperforms the proposed CNN-MLP, CNN-ENN and CNN-FENN methods and improved the classification performance metrics across 5-fold cross-validation with an average sensitivity of 79%, accuracy of 78% and specificity of 76%. The proposed methods can reduce the likelihood of hazardous events like pulmonary congestion and ventricular suction for HF patients and notify identified abnormal cases to the hospital, clinician and cardiologist for emergency action, which can diminish the mortality rate of patients with HF.

A Sensorless Control System for an Implantable Heart Pump Using a Real-Time Deep Convolutional Neural Network

Left ventricular assist devices (LVADs) are mechanical pumps, which can be used to support heart failure (HF) patients as bridge to transplant and destination therapy. To automatically adjust the LVAD speed, a physiological control system needs to be designed to respond to variations of patient hemodynamics across a variety of clinical scenarios. These control systems require pressure feedback signals from the cardiovascular system. However, there are no suitable long-term implantable sensors available. In this study, a novel real-time deep convolutional neural network (CNN) for estimation of preload based on the LVAD flow was proposed. A new sensorless adaptive physiological control system for an LVAD pump was developed using the full dynamic form of model free adaptive control (FFDL-MFAC) and the proposed preload estimator to maintain the patient conditions in safe physiological ranges. The CNN model for preload estimation was trained and evaluated through 10-fold cross validation on 100 different patient conditions and the proposed sensorless control system was assessed on a new testing set of 30 different patient conditions across six different patient scenarios. The proposed preload estimator was extremely accurate with a correlation coefficient of 0.97, root mean squared error of 0.84 mmHg, reproducibility coefficient of 1.56 mmHg, coefficient of variation of 14.44%, and bias of 0.29 mmHg for the testing dataset. The results also indicate that the proposed sensorless physiological controller works similarly to the preload-based physiological control system for LVAD using measured preload to prevent ventricular suction and pulmonary congestion. This study shows that the LVADs can respond appropriately to changing patient states and physiological demands without the need for additional pressure or flow measurements.

Improving In vitro Evaluation Capabilities of Cardiac Assist Devices through a Validated Exercise Simulation

Cardiac assist devices require thorough in vitro evaluation prior to in vivo animal trials, which is often undertaken in mock circulatory loops. To allow for best possible device development, mock circulatory loops need to be able to simulate a variety of patient scenarios. Transition from rest to exercise is one of the most commonly simulated patient scenarios, however, to validate in vitro exercise test beds, baseline data on how the healthy heart and circulatory system responds to exercise is required. Steady state and time response data for heart rate (HR), stroke volume (SV) and cardiac output (CO) was continuously recorded using impedance cardiography in 50 healthy subjects (27 male / 23 female) during exercise on a recumbent exercise ergometer. This data was then used to implement an exercise simulation in a mock circulatory loop and both the steady state and transient results were compared with the mean response of subjects transitioning from rest to 60 W exercise. When transitioning from rest to exercise the time constant (τ) and rise time (t_r) for HR, SV and CO were between 10.6-19.3s and 24.7-44.3s respectively for both sexes. No significant differences between the genders were found for τ and t_r (p>0.05). Mock circulatory loop results of HR, SV and CO were in good accordance with human data. The present data was used to successfully validate in vitro exercise simulations and may be used to validate in silico numerical simulations of exercise, thus further improving the evaluation capabilities for existing and under development cardiac assist devices.

A Starling-like total work controller for rotary blood pumps: An in vitro evaluation

Due to improved durability and survival rates, rotary blood pumps (RBPs) are the preferred left ventricular assist device when compared to volume displacement pumps. However, when operated at constant speed, RBPs lack a volume balancing mechanism which may result in left ventricular suction and suboptimal ventricular unloading. Starling-like controllers have previously been developed to balance circulatory volumes; however, they do not consider ventricular workload as a feedback and may have limited sensitivity to adjust RBP workload when ventricular function deteriorates or improves. To address this, we aimed to develop a Starling-like total work controller (SL-TWC) that matched the energy output of a healthy heart by adjusting RBP hydraulic work based on measured left ventricular stroke work and ventricular preload. In a mock circulatory loop, the SL-TWC was evaluated using a HeartWare HVAD in a range of simulated patient conditions. These conditions included changes in systemic hypertension and hypotension, pulmonary hypertension, blood circulatory volume, exercise, and improvement and deterioration of ventricular function by increasing and decreasing ventricular contractility. The SL-TWC was compared to constant speed control where RBP speed was set to restore cardiac output to 5.0 L/min at rest. Left ventricular suction occurred with constant speed control during pulmonary hypertension but was prevented with the SL-TWC. During simulated exercise, the SL-TWC demonstrated reduced LVSW (0.51 J) and greater RBP flow (9.2 L/min) compared to constant speed control (LVSW: 0.74 J and RBP flow: 6.4 L/min). In instances of increased ventricular contractility, the SL-TWC reduced RBP hydraulic work while maintaining cardiac output similar to the rest condition. In comparison, constant speed overworked and increased cardiac output. The SL-TWC balanced circulatory volumes by mimicking the Starling mechanism, while also considering changes in ventricular workload. Compared to constant speed control, the SL-TWC may reduce complications associated with volume imbalances, adapt to changes in ventricular function and improve patient quality of life.

Preload Sensitivity with TORVAD Counterpulse Support Prevents Suction and Overpumping

PURPOSE

This study compares preload sensitivity of continuous flow (CF) VAD support to counterpulsation using the Windmill toroidal VAD (TORVAD). The TORVAD is a two-piston rotary pump that ejects 30 mL in early diastole, which increases cardiac output while preserving aortic valve flow.

METHODS

Preload sensitivity was compared for CF vs. TORVAD counterpulse support using two lumped parameter models of the cardiovascular system: (1) an open-loop model of the systemic circulation was used to obtain ventricular function curves by isolating the systemic circulation and prescribing preload and afterload boundary conditions, and (2) a closed-loop model was used to test the physiological response to changes in pulmonary vascular resistance, systemic vascular resistance, heart rate, inotropic state, and blood volume. In the open-loop model, ventricular function curves (cardiac output vs left ventricular preload) are used to assess preload sensitivity. In the closed-loop model, left ventricular end systolic volume is used to assess the risk of left ventricular suction.

RESULTS

At low preloads of 5 mmHg, CF support overpumps the circulation compared to TORVAD counterpulse support (cardiac output of 3.3 L/min for the healthy heart, 4.7 with CF support, and 3.5 with TORVAD counterpulse support) and has much less sensitivity than counterpulse support (0.342 L/min/mmHg for the healthy heart, 0.092 with CF support, and 0.306 with TORVAD counterpulse support). In the closed-loop model, when PVR is increased beyond 0.035 mmHg s/mL, CF support overpumps the circulation and causes ventricular suction events, but TORVAD counterpulse support maintains sufficient ventricular volume and does not cause suction.

CONCLUSIONS

Counterpulse support with the TORVAD preserves aortic valve flow and provides physiological sensitivity across all preload conditions. This should prevent overpumping and minimize the risk of suction.

A Physiological Control System for an Implantable Heart Pump That Accommodates for Interpatient and Intrapatient Variations

Left ventricular assist devices (LVADs) can provide mechanical support for a failing heart as a bridge to transplant and destination therapy. Physiological control systems for LVADs should be designed to respond to changes in hemodynamic across a variety of clinical scenarios and patients by automatically adjusting the heart pump speed. In this study, a novel adaptive physiological control system for an implantable heart pump was developed to respond to interpatient and intrapatient variations to maintain the left-ventricle-end-diastolic-pressure (LVEDP) in the normal range of 3 to 15 mmHg to prevent ventricle suction and pulmonary congestion. A new algorithm was also developed to detect LVEDP from pressure sensor measurement in real-time mode. Model-free adaptive control (MFAC) was employed to control the pump speed via simulation of 100 different patient conditions in each of six different patient scenarios, and compared to standard PID control. Controller performance was tracked using the sum of the absolute error (SAE) between the desired and measured LVEDP. The lower SAE on control tracking performance means that the measured LVEDP follows the desired LVEDP faster and with less amplitude oscillations, preventing ventricle suction and pulmonary congestion (mean and standard deviation of SAE (mmHg) for all 600 simulations were 18813 ± 29345 and 24794 ± 28380 corresponding to MFAC and PID controller, respectively). In four out of six patient scenarios, MFAC control tracking performance was better than the PID controller. This study shows the control performance can be guaranteed across different patients and conditions when using MFAC over PID control, which is a step toward clinical acceptance of these systems.

The Importance of Venous Return in Starling-Like Control of Rotary Ventricular Assist Devices

Rotary ventricular assist devices (VADs) are less sensitive to preload than the healthy heart, resulting in inadequate flow regulation in response to changes in patient cardiac demand. Starling-like physiological controllers (SLCs) have been developed to automatically regulate VAD flow based on ventricular preload. An SLC consists of a cardiac response curve (CRC) which imposes a nonlinear relationship between VAD flow and ventricular preload, and a venous return line (VRL) which determines the return path of the controller. This study investigates the importance of a physiological VRL in SLC of dual rotary blood pumps for biventricular support. Two experiments were conducted on a physical mock circulation loop (MCL); the first compared an SLC with an angled physiological VRL (SLC-P) against an SLC with a vertical VRL (SLC-V). The second experiment quantified the benefit of a dynamic VRL, represented by a series of specific VRLs, which could adapt to different circulatory states including changes in pulmonary (PVR) and systemic (SVR) vascular resistance versus a fixed physiological VRL which was calculated at rest. In both sets of experiments, the transient controller responses were evaluated through reductions in preload caused by the removal of fluid from the MCL. The SLC-P produced no overshoot or oscillations following step changes in preload, whereas SLC-V produced 0.4 L/min (12.5%) overshoot for both left and right VADs. Additionally, the SLC-V had increased settling time and reduced controller stability as evidenced by transient controller oscillations. The transient results comparing the specific and standard VRLs demonstrated that specific VRL rise times were improved by between 1.2 and 4.7 s ( x ¯ = 3.05 s), while specific VRL settling times were improved by between 2.8 and 16.1 seconds ( x ¯ = 8.38 s) over the standard VRL. This suggests only a minor improvement in controller response time from a dynamic VRL compared to the fixed VRL. These results indicate that the use of a fixed physiologically representative VRL is adequate over a wide variety of physiological conditions.

A Multiphysics Biventricular Cardiac Model: Simulations With a Left-Ventricular Assist Device

Computational models have become essential in predicting medical device efficacy prior to clinical studies. To investigate the performance of a left-ventricular assist device (LVAD), a fully-coupled cardiac fluid-electromechanics finite element model was developed, incorporating electrical activation, passive and active myocardial mechanics, as well as blood hemodynamics solved simultaneously in an idealized biventricular geometry. Electrical activation was initiated using a simplified Purkinje network with one-way coupling to the surrounding myocardium. Phenomenological action potential and excitation-contraction equations were adapted to trigger myocardial contraction. Action potential propagation was formulated within a material frame to emulate gap junction-controlled propagation, such that the activation sequence was independent of myocardial deformation. Passive cardiac mechanics were governed by a transverse isotropic hyperelastic constitutive formulation. Blood velocity and pressure were determined by the incompressible Navier-Stokes formulations with a closed-loop Windkessel circuit governing the circulatory load. To investigate heart-LVAD interaction, we reduced the left ventricular (LV) contraction stress to mimic a failing heart, and inserted a LVAD cannula at the LV apex with continuous flow governing the outflow rate. A proportional controller was implemented to determine the pump motor voltage whilst maintaining pump motor speed. Following LVAD insertion, the model revealed a change in the LV pressure-volume loop shape from rectangular to triangular. At higher pump speeds, aortic ejection ceased and the LV decompressed to smaller end diastolic volumes. After multiple cycles, the LV cavity gradually collapsed along with a drop in pump motor current. The model was therefore able to predict ventricular collapse, indicating its utility for future development of control algorithms and pre-clinical testing of LVADs to avoid LV collapse in recipients.

Synergy of first principles modelling with predictive control for a biventricular assist device: In silico evaluation study

Control for dual rotary left ventricular assist devices (LVADs) used as a biventricular assist device (BiVAD) is challenging. If the control system fails, flow imbalance between the systemic and the pulmonary circulations would result, subsequently leading to ventricular suction or pulmonary congestion. With the expectation that advanced control approaches such as model predictive control could address the challenges naturally and effectively, we developed a synergistic first principles model predictive controller (MPC) for the BiVAD. The internal model of the MPC is a simplified state-space model that has been developed and validated in a previous study. A single Frank-Starling (FS) control curve was used to define the target pump flow corresponding to the preload on each side of the heart. The MPC was evaluated in a validated numerical model using three clinical scenarios: blood loss, myocardial recovery, and exercise. Simulation results showed that the MPC was effective in adapting to changes in physiological states without causing ventricular suction or pulmonary congestion. The use of MPC for a BiVAD eliminates the need for two controllers of dual LVADs thus making the task of controller tuning easier.

Sensor-Based Physiologic Control Strategy for Biventricular Support with Rotary Blood Pumps

Rotary biventricular assist devices (BiVAD) are becoming a clinically accepted treatment option for end-stage biventricular failure. To improve BiVAD efficacy and safety, we propose a control algorithm to achieve the clinical objectives of maintaining left-right-sided balance, restoring physiologic flows, and preventing ventricular suction. The control algorithm consists of two proportional-integral (PI) controllers for left and right ventricular assist devices (LVAD and RVAD) to maintain differential pump pressure across LVAD (ΔPL) and RVAD (ΔPR) to provide left-right balance and physiologic flow. To prevent ventricular suction, LVAD and RVAD pump speed differentials (ΔRPML, ΔRPMR) were maintained above user-defined thresholds. Efficacy and robustness of the proposed algorithm were tested in silico for axial and centrifugal flow BiVAD using 1) normal and excessive ΔPL and/or ΔPR setpoints, 2) rapid threefold increase in pulmonary vascular or vena caval resistances, 3) transient responses from exercise to rest, and 4) ventricular fibrillation. The study successfully demonstrated that the proposed BiVAD algorithm achieved the clinical objectives but required pressure sensors to continuously measure ΔPL and ΔPR. The proposed control algorithm is device independent, should not require any modifications to the pump or inflow/outflow cannulae/grafts, and may be directly applied to current rotary blood pumps for biventricular support.

In-vitro evaluation of physiological controller response of rotary blood pumps to changes in patient state

Rotary blood pumps (RBPs) have a low sensitivity to preload changes when run at constant speed, which can lead to harmful ventricular suction events. Therefore a control mechanism is needed to adjust RBP speed in response to patient demand, but an appropriate response time for physiological control strategies to these changes in patient demand has not been determined. This paper aims to evaluate the response of a simulated healthy heart with those of different RBP control techniques during exercise simulations and a Valsalva manoeuver. A mock circulation loop was used to simulate the response of a healthy heart to these changes in patient state. The generated data was compared with a simulated RBP (VentrAssist) supported left heart failure condition. A range of control techniques including constant speed, proportional integral (PI) (active control) and a compliant inflow cannula (passive control) were used to achieve restored haemodynamics and evaluate controller response time. Controllers that responded faster (active control) or slower (active control and constant speed mode) than the native heart's response led to ventricular suction. Active control systems can respond both faster or slower than the heart depending on the controller gains. A control system that responded similar to the native heart was able to prevent ventricular suction. This study concluded that a physiological control system should mimic the response of the native heart to changes in patient state in order to prevent ventricular suction.