Multiple Model Predictive Control of the Cardiovascular System using Vagal Nerve Stimulation

Area of Science:

• Cardiovascular Medicine
• Biomedical Engineering
• Control Systems

Background:

• Vagal nerve stimulation (VNS) shows potential for treating cardiovascular diseases like heart failure, arrhythmia, and hypertension.
• Current VNS therapies use open-loop systems with heuristically determined parameters, leading to inconclusive efficacy.
• A closed-loop approach is needed to optimize patient-specific VNS parameters for improved therapeutic outcomes.

Purpose of the Study:

• To develop a closed-loop multiple model predictive control (MPC) algorithm for automated VNS.
• To optimize stimulation parameters (amplitude and frequency) for regulating heart rate and mean arterial pressure.
• To validate the computational efficiency of the closed-loop VNS algorithm using hardware-in-the-loop simulation.

Main Methods:

• Developed a multiple model predictive control algorithm for VNS.
• Utilized a previously reported pulsatile rat cardiac model simulating hypertension.
• Identified local models from the cardiac model for rest and exercise states.
• Adjusted electrical pulse amplitude and frequency at three vagal nerve locations.
• Verified computational expense via hardware-in-the-loop implementation.

Main Results:

• Successfully developed and simulated a closed-loop VNS control algorithm.
• Demonstrated the algorithm's capability to regulate heart rate and mean arterial pressure.
• Validated the computational feasibility of the proposed MPC approach.

Conclusions:

• A closed-loop MPC algorithm offers a promising approach for optimizing VNS therapy.
• Automated, patient-specific VNS parameter adjustment can enhance treatment efficacy for cardiovascular conditions.
• The developed method shows potential for clinical translation in managing heart rate and blood pressure.