Intelligent Control System applied to In Vitro Ventricular Assist Device Test Bench for Decision Support

The Ventricular Assist Device (VAD) is a mechatronic device used as a therapy for advanced heart failure (HF) in patients who are candidates for organ transplantation. Although the most frequent application is still a short-term (transplant bridge) therapy strategy, long-term use (destination therapy) has been increasing because of the increased longevity of the population using these devices. However, according to the INTERMACS report, adverse events are observed that compromise the survival rate of implanted patients, among which we have the malfunction of the VAD. The occurrence of this event influences the reliability and availability of the device. In this way, the development of resources and techniques to improve the robustness, availability and reliability of these devices is indispansable for the medical equipment industry. This prospective work proposes an in vitro test bench with intelligent behavior, which allows to put of VAD in operation. The test bench has rotation, vibration, energy, temperature, pressure and flow sensors that are monitored by a computer system. In order to identify malfunctions, standard curves have been developed for the variables being monitored. When a deviation of a variable with respect to its standard curve appears, a fault is characterized. The collection, treatment and monitoring of this failure data is accomplished by the use of Big Data Analytics technology. The method proposed in this work allows the verification and validation of a VAD, based on the behavioral data collected during the tests, through the appropriate protocols and the in vitro procedures.