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

Area of Science:

• Biomedical Engineering
• Cardiovascular Engineering
• Control Systems

Background:

• Controlling dual rotary left ventricular assist devices (LVADs) as a biventricular assist device (BiVAD) presents significant challenges.
• Failure in BiVAD control can lead to critical flow imbalances between systemic and pulmonary circulations, causing ventricular suction or pulmonary congestion.

Purpose of the Study:

• To develop and evaluate a synergistic first principles model predictive controller (MPC) for BiVADs.
• To address the inherent control challenges in BiVADs and enhance patient safety.

Main Methods:

• Developed a synergistic first principles model predictive controller (MPC) utilizing a simplified state-space model.
• Employed a single Frank-Starling (FS) control curve to regulate target pump flow based on cardiac preload.
• Evaluated the MPC in a validated numerical model under clinical scenarios: blood loss, myocardial recovery, and exercise.

Main Results:

• The MPC effectively adapted to physiological changes without inducing ventricular suction or pulmonary congestion.
• Simulations demonstrated the controller's capability to maintain stable circulation during various clinical conditions.
• The developed MPC successfully managed BiVAD function, preventing adverse hemodynamic events.

Conclusions:

• The synergistic first principles MPC offers an effective solution for BiVAD control.
• This approach simplifies BiVAD management by eliminating the need for dual LVAD controllers, easing controller tuning.
• The MPC demonstrates potential for improving safety and efficacy in BiVAD applications.