Model predictive control of depth of anaesthesia: guidelines for controller configuration

Area of Science:

• Anesthesiology and Critical Care
• Control Systems Engineering
• Pharmacology

Background:

• Accurate regulation of Depth of Anesthesia (DoA) is crucial for patient safety during surgery.
• Existing control methods may not fully address the complexities of anesthetic drug pharmacodynamics.
• The Bispectral Index (BIS) is a common measure for monitoring DoA.

Purpose of the Study:

• To investigate the feasibility of Model Predictive Control (MPC) for regulating DoA.
• To develop and evaluate an MPC strategy using linearized pharmacokinetic/pharmacodynamic (PK/PD) models.
• To characterize the impact of MPC design parameters on DoA control performance.

Main Methods:

• Linearization of nonlinear PK/PD models for Propofol and Remifentanil around a target DoA working point.
• Development of a linear state-space model incorporating integral action for enhanced control.
• Implementation of the linear model within a Model Predictive Control (MPC) framework.

Main Results:

• Successful application of MPC for regulating DoA based on BIS index measurements.
• Demonstration of the influence of various MPC design parameters on control accuracy and stability.
• Validation of the linearized model approach for MPC in anesthesia control.

Conclusions:

• Model Predictive Control (MPC) is a feasible approach for automated anesthesia depth regulation.
• The study provides insights into optimizing MPC design for clinical application in anesthesia.
• Further research can refine MPC strategies for improved patient outcomes and drug titration.