In vitro evaluation of multi‐objective physiological control of the centrifugal blood pump

Abstract

In recent years, left ventricular assist devices (LVADs) have been successfully used as a bridge to heart transplantation or as destination therapy (DT) for the treatment of congestive heart failure (CHF). Continuous flow LVADs are smaller, more reliable, and less complex than the first generation LVADs (pulsatile). 1-4 The development of control systems which are able to adapt according to the body's metabolic demands is called physiological control. Research in this field has already been done since the early 1990s. 5-8 The purpose of LVAD control

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

“…In contrast, in this study, the preload was estimated using deep CNNs based on the pump flow and FFDL-MFAC was used to identify the full dynamics of the CVS and LVAD and control the pump speed. Moreover, the use of an FFDL-MFAC controller in this study is advantageous because it does not have the common issues of other adaptive controllers such as ANN [17] and fuzzy controllers [20]. These controllers generate a high computational burden and rely on accurate rules for different patient conditions, respectively.…”

“…Furthermore, in most studies of physiological control systems for LVADs [13]- [15], [18], a proportionalintegral-derivative (PID) controller was employed, which was 2 tuned for a specific patient and condition, and control performance cannot be guaranteed across different patients and conditions, which may lead to hazardous events [19]. In recent studies, adaptive control systems such as artificial neural network (ANN) control [17] and fuzzy logic control (FLC) [20], which can automatically adjust their parameters according to feedback from the controlled system, have been used in the control of LVADs. However, these controllers need large training data and precise determination of rules for different patient conditions, respectively.…”

Latin American Society for Artificial Organs, Biomaterials, and Tissue Engineering: Update 2020

“…3 Leao and colleagues developed a control system/ algorithm consisting of a control system that harmoniously adjusts pump speed without additional sensors, considering the patient's clinical condition and physical activity. 4 Two additional papers dealt with materials and biocompatibity. Cappelato et al used human dermal fibroblasts to evaluate the electrospinning process of the biocompatible polymer (PCL) with TiO 2 nanotubes on the Ti-30Ta alloy surface.…”

“…An article by de Souza et al presents the design strategy for development of a customized ventricular assist device electromagnetic actuator 3 . Leao and colleagues developed a control system/algorithm consisting of a control system that harmoniously adjusts pump speed without additional sensors, considering the patient's clinical condition and physical activity 4 …”

Control Strategy Design of a Microblood Pump Based on Heart-Rate Feedback

“…Petrou et al [ 28 ] developed a multi-objective physiological control system that was dependent on pump inlet pressure (PIP) and implemented signal processing algorithms to extract features from PIP that can meet various objectives; however, the control method relied on the development of a blood-pressure sensor with good blood compatibility. Leao et al [ 29 ] proposed a multi-objective physiological control system based on fuzzy logic to maintain the mean arterial pressure and cardiac output at the physiological level.…”