Tuning of Hypnosis Controller from Data During Anesthesia Induction Using FRIT

Abstract

Abstract Many studies of automatic anesthesia control systems to maintain an appropriate anesthetic state during surgery have been done. Since individual difference in the response of a patient to anesthetic drugs is quite large, the controller must be adjusted for each patient to achieve an appropriate anesthesia control. In our pilot study, we applied the fictitious reference iterative tuning (FRIT) method to tuning of a proportional-integral-derivative (PID) controller in a hypnosis control system for in silico patients with individual differences in gain and another parameter and obtained an appropriate controller from 7-minute data during anesthesia induction in most cases. In this paper, we apply the method to in silico patients with individual differences in pharmacodynamic parameters and dead-time, and evaluate the effectiveness of the method.