Yih-Choung Yu

Estimation of systemic vascular bed parameters for artificial heart control

An extended Kalman filter estimator for the identification of systemic circulation model parameters during cardiac ejection and cardiac filling is described. The estimator has been developed for use in the control of a cardiac ventricular assist device. A lumped element circuit with a time-varying capacitor was used to represent the systemic circulation and the left ventricle. Since the haemodynamic variables that are measurable in patients with impaired cardiac function vary dramatically as the patients move through different levels of care, the estimator was designed so that it can be used with different sets of blood pressure and flow measurements. Preliminary evaluation of the performance of the estimator using data from a computer simulation and from a patient during open-heart surgery is presented. The robustness of the estimator to variations in parameter initialization is also described.

Intelligent control design for heart assist devices

Heart assist devices are blood pumps used to augment the cardiac output of patients with left ventricular failure. A new generation of devices being evaluated for human use is based on turbo-hydrodynamic methods of pumping, which offer several advantages over the reciprocating, pulsatile methods used in current devices. However, the new devices pose a more difficult control problem because of their sensitivity to circulatory load and other patient cardiovascular parameters. The paper describes the design of a control structure to regulate the operation of these devices. The controller has three different types of algorithm available: a model-based patient-adaptive algorithm; two heuristic algorithms that rely only on the device characteristics; and a default algorithm. The patient-adaptive algorithm uses a model of the patients systemic circulation to determine the required cardiac output for a given level of activity. The heuristic algorithms use the known operating characteristics of the device to adjust the cardiac output to changes in demand without knowledge of patient-specific conditions. The default algorithm provides a fixed speed operation to be used in case of system or sensor failure. An intelligent supervisor determines the cardiac output required from the assist device and selects the control algorithm to use, based on a multidimensional measure of the patients level of activity, available estimates of hemodynamic variables, reliability of the patient model, and the past history of the patient.

Minimally Invasive Estimation of Systemic Vascular Parameters

AbstractA cardiovascular parameter estimator to identify the systemic vascular parameters was developed using an extended Kalman filter (EKF) algorithm. Measurements from a ventricular assist device (VAD) and arterial pressure were used in the estimator. The systemic vascular parameters are important indices of heart condition. However, obtaining these parameters usually requires invasive measurements, which are difficult to obtain under most clinical environments. Including a VAD model into the estimator and using the signals from a VAD to identify the cardiovascular parameters for VAD patients would minimize the need for indwelling sensors. This paper illustrates the use of a Novacor left ventricular assist system (LVAS) model with a cardiovascular model in the estimator to identify the systemic vascular parameters: characteristic resistance, blood inertance at the aorta, systemic compliance, and systemic resistance. Performance of the estimator was evaluated using data from a computer simulation and from a mock circulatory system experiment. Robustness of the estimator to the available measurements was also described. The estimation results showed that the estimates converged with reasonable accuracy in a limited time when the LVAS pump volume and arterial pressure were used as measurements. These parameter estimates can provide additional diagnostic information for patient and device monitoring and can be used for future VAD control development.

Control issues in rotary heart assist devices

Heart assist devices are mechanical pumps used in patients with cardiovascular diseases who are awaiting heart transplantation. With increasing clinical success, these devices are being used for longer periods of time, and automatic control has become an important requirement. The principal control requirement is to mimic the normal response of the heart to changes in demand for cardiac output, and control therefore provides many challenging problems for control engineers involved in this important multidisciplinary area of research. A team of engineers and physicians at the University of Pittsburgh is developing a hierarchical structure for a control system to support a rotary-type heart assist device. In this paper, we review various types of devices, the control algorithms used, and the problems that must be solved.

Performance Prediction of a Percutaneous Ventricular Assist System Using Nonlinear Circuit Analysis Techniques

A percutaneous ventricular assist device (pVAD) is an extracorporeal cardiac assist system that supports the failing ventricle in advanced stage heart failure by bypassing blood from the venous to the arterial circulation through a blood pump. The system can be implanted in a Cath lab using standard interventional techniques, and typically consists of a venous or atrial drainage cannula, the VAD (or blood pump), and an arterial perfusion cannula. Because the device allows clinicians the freedom of choosing the configuration and size of the cannulae based on the patients body size and the size of the artery, it is extremely difficult but important to be able to predict the amount of blood flow that the device can provide before it is implanted to support the patient. In this paper, we develop a novel method that can be used to accurately predict the mean flow rate that the device can provide to the patient based on the size and configuration of the arterial cannula, the pump speed, and the patients left atrial and mean arterial pressures. To do this, we first develop a nonlinear electric circuit model for the pVAD. This model includes a speed dependent voltage source and flow dependent resistors to simulate the pressure-flow relationship in the various cannulae in the device. We show that the flow rate through the device can be determined by solving a quadratic equation whose coefficients are scaled depending on the size and configuration of the arterial cannula. The model and prediction method were tested experimentally on a test loop supported by the TandemHeart pVAD (Cardiacassist, Inc., Pittsburgh, PA). A comparison of the predicted flow rates obtained from our method with experimental data shows that our method can predict the flow rates accurately with error indices less than 6% for all test conditions over the entire range of intended use of the device. Computer simulations of the pVAD model coupled to a cardiovascular model showed that the accuracy of the method in estimating the mean flow rate is consistent over the normal range of operation of the device regardless of the pulsatility introduced by the cardiovascular system. This method can be used as an additional too to assist cardiologists in choosing a proper arterial cannulae configurations and sizes for pVAD patients. It can also be used as a tool to train clinical personnel to operate the device under different physiological conditions.

Minimally invasive estimation of systemic vascular parameters for artificial heart control

Abstract Systemic vascular parameters are important indices of heart condition, incorporating these parameters into the control of a ventricular assist device (VAD) would facilitate the implementation of an effective control strategy. In order to minimize the need for indwelling sensors for obtaining these parameters, an estimator was developed to identify the systemic vascular parameters (characteristic resistance, blood inertance at the aorta, systemic compliance, and systemic resistance) using measurements from the Novacor left ventricular assist system (LVAS) and arterial pressure. Systemic compliance was estimated by the ratio of the LVAS pump stroke volume to the arterial pulse pressure and systemic resistance was calculated by the ratio of mean arterial pressure to LVAS pump output. These two parameters were then used as known parameters in an extended Kalman filter to identify the other unknown parameters using LVAS pump volume and arterial pressure measurements. Performance of the estimator was evaluated using data from a mock circulatory system experiment. The results showed that the estimates converged more accurately in a limited time when arterial pressure was used with the LVAS pump volume as measurements. These parameter estimates can provide diagnostic information for patient and device monitoring and can be used for future VAD control development.

Pressure-volume relationship of a pulsatile blood pump for ventricular assist device development

A mathematical model describing the pressure-volume relationship of the Novacor left ventricular assist system (LVAS) was developed. The model consists of lumped resistance, capacitance, and inductance elements with one time varying capacitor to estimate the cyclic pressure generation of the pump using pump volume measurement. The ejection and filling portions of the pump cycle were modeled with two separate functions. The corresponding model parameters were estimated by least squares fit to experimental data obtained in the laboratory. Pressure and volume waveforms obtained from the model were compared with data obtained from laboratory tests and from patients. It performed well in simulating pump operation throughout the entire cycle. This model can be used for the evaluation of LVAS performance, for on-line estimation of an LVAS patient’s cardiovascular parameters, for pump controller development, and as a tool for engineer training.

Mathematical modeling of ventricular suction induced by a rotary ventricular assist device

Ventricular suction (VS) is an event caused by operating a rotary ventricular assist device (VAD) at a high speed in a patient with low blood volume in the left ventricle (LV). VS occurs when the blood flow out of the ventricle exceeds the flow into the ventricle, causing the ventricle to collapse. Because of this, it is risky to use a VAD on a long-term basis without changing the speed as the demand of a human body changes dramatically over time. As a result, a controller that can automatically adjust pump speed based on a patients physiological needs with the capability of suction detection and avoidance would be important for developing a rotary VAD for long-term use. Computer modeling of the cardiovascular system has been used extensively for controller design and testing. To test the capability of a VAD controller to respond to VS, a mathematical model that can simulate VS must be integrated with a cardiovascular system model. In this paper, a nonlinear resistance as a function of the LV pressure, the absolute value of the LV pressure derivative, and the pump inlet pressure was identified to model VS. This was done by fitting a curve to data sets from animal experiments with the HeartMate II VAD. The model fit the data well with an average root-mean-squared (RMS) error of 10.93%. The model was validated against 23 different animal data sets with a mean RMS error of 29.98%. An existing suction model was tested with the same validation data sets and yielded a mean RMS error of 67.41%. The newly developed model showed a significant improvement from the existing model. It will be integrated with a cardiovascular model for future development of a suction detector and physiologic controller for a VAD

Sensitivity analysis of cardiovascular models for minimally invasive estimation of systemic vascular parameters

A sensitivity analysis technique was used to evaluate two simple cardiovascular models for model parameter estimation using minimal measurable signals. The purpose of this study is to determine a simple model, which can used with a left ventricular device (LVAD) model to identify the model parameters using signals from the LVAD, to characterize the systemic load. The control of the LVAD can then be adjusted based on patients demand, represented by the systemic load, with minimum invasive sensors. The results obtained from the nominal parameter values of normal patients suggested that a model with a constant capacitance for the left atrium is suitable for parameter estimation if aortic pressure, aortic flow, and left atrial pressure are measurable. If only aortic pressure and flow are available, a simplified model, excluding the left atrium from the original model, provides more accurate estimates. An extended Kalman filter was used with computer simulation data to identify parameters of the models for verifying the results from the sensitivity analysis.

Minimally invasive estimation of cardiac function for patients with rotary VAD support

Maximum ventricular elastance, E/sub MAX/, is a reliable quantitative index representing the contractual status of a patients heart. However evaluating E/sub MAX/ usually requires invasive pressure and flow sensors, which only can be performed under certain clinical facility. If an indirect index of E/sub MAX/ can be identified from the measurements of a ventricular assist device (VAD) without any indwelling sensor, this would facilitate an effective way to monitor the healthy condition of the patients heart while the patients are under VAD support. This index can also be used to determine the control strategy of VAD operation and gradually wean the patient from the mechanical circulatory support. In this paper, two possible indices, pump flow pulsatility and arterial pulse pressure, were evaluated as alternative representations of E/sub MAX/ using data from a computer simulation model of cardiovascular system with a HeartMat II left ventricular assist system (LVAS). Pump flow pulsatility showed a strong correlation to E/sub MAX/ regardless of pump speed changes, and thus can be used as an index for E/sub MAX/.