First-Hand Design of a Fractional order PID for Controlling the Depth of Hypnosis during Induction

Abstract

The integration of information technology and control engineering has significantly increased in the research field of clinical practice, including the management of medication dosing for general anesthesia. To achieve effective drug dosing, it is necessary to have appropriate controllers for closed loop control methods. A multitude of control mechanisms have been devised to manage hypnosis, with most of them based on pharmacokinetic-pharmacodynamic (PK-PD) models. This study presents a novel structure of a fractional order PID controller tuned for inducing hypnosis. The tuning algorithm uses a novel fractional order model instead of the classical PK-PD models. The main goal is to control the Bispectral Index by delivering Propofol during the hypnotic stage of anesthesia. Closed loop simulation results show that the proposed controller manages to ensure little undershoot and time to target for a nominal patient model. Robustness tests are also performed. The analysis shows that the proposed controller is suitable to maintain BIS signal within a safe operating range but fails to meet the time to target requirement for significant patient variability (greater than 2 orders of magnitude). Online adaptation rules are suggested as a solution.