Evaluation of a long-range adaptive predictive controller for computerized drug delivery systems

Area of Science:

• Biomedical Engineering
• Control Systems
• Pharmacology

Background:

• Computerized drug delivery systems require precise control for patient safety and therapeutic efficacy.
• Adaptive control strategies are crucial for managing physiological variables like mean arterial pressure (MAP) in dynamic environments.

Purpose of the Study:

• To develop and evaluate a closed-loop adaptive control system for computerized drug delivery.
• To assess the system's ability to regulate mean arterial pressure (MAP) using sodium nitroprusside infusion.

Main Methods:

• Implementation of a generalized predictive control law with a terminal matching condition for drug infusion.
• Utilizing a control-relevant, long-range identification algorithm for on-line parameter estimation.
• Experimental validation in mongrel dogs to evaluate setpoint tracking and disturbance rejection of MAP.

Main Results:

• The system achieved hypotension induction within an average of 2.44 +/- 0.31 minutes without MAP overshoots.
• Mean arterial pressure remained within 5 mm Hg of the target 96.2% of the time under stable conditions.
• The control system demonstrated robustness and effectiveness when subjected to unpredictable disturbances.

Conclusions:

• The developed closed-loop adaptive control system is effective and robust for computerized drug delivery.
• The system accurately regulates mean arterial pressure, even in the presence of physiological constraints and disturbances.
• This approach offers a promising solution for automated and precise drug administration in clinical settings.