One of the most important recent improvements in cardiology is the use of ventricular assist devices (VADs) to help patients with severe heart diseases, especially when they are indicated to heart transplantation. The Institute Dante Pazzanese of Cardiology has been developing an implantable centrifugal blood pump that will be able to help a sick human heart to keep blood flow and pressure at physiological levels. This device will be used as a totally or partially implantable VAD. Therefore, an improvement on device performance is important for the betterment of the level of interaction with patient's behavior or conditions. But some failures may occur if the device's pumping control does not follow the changes in patient's behavior or conditions. The VAD control system must consider tolerance to faults and have a dynamic adaptation according to patient's cardiovascular system changes, and also must attend to changes in patient conditions, behavior, or comportments. This work proposes an application of the mechatronic approach to this class of devices based on advanced techniques for control, instrumentation, and automation to define a method for developing a hierarchical supervisory control system that is able to perform VAD control dynamically, automatically, and securely. For this methodology, we used concepts based on Bayesian network for patients' diagnoses, Petri nets to generate a VAD control algorithm, and Safety Instrumented Systems to ensure VAD system security. Applying these concepts, a VAD control system is being built for method effectiveness confirmation.
© 2011, Copyright the Authors. Artificial Organs © 2011, International Center for Artificial Organs and Transplantation and Wiley Periodicals, Inc.