Robust Predictive Control Strategy Applied for Propofol Dosing Using BIS as a Controlled Variable During Anesthesia

Abstract

This paper presents the application of predictive control to drug dosing during anesthesia in patients undergoing surgery. The performance of a generic predictive control strategy in drug dosing control, with a previously reported anesthesia-specific control algorithm, has been evaluated. The robustness properties of the predictive controller are evaluated with respect to inter- and intrapatient variability. A single-input (propofol) single-output (bispectral index, BIS) model of the patient has been assumed for prediction as well as for simulation. A set of 12 patient models were studied and interpatient variability and disturbances are used to assess robustness of the controller. Furthermore, the controller guarantees the stability in a desired range. The applicability of the predictive controller in a real-life environment via simulation studies has been assessed.

Positive state observer for the automatic control of the depth of anesthesia—Clinical results

“…However, due to the high complexity of this procedure an automated system for drug administration would be a good support for the clinicians (see Meijler [6]). The development of controllers for the automatic administration of drugs in patients has deserved the attention of several researchers and led to a number of contributions and controllers namely a predictive control in Ionescu et al [7], an adaptive model-based controller in Mortier et al [8] and Simanski et al [9], a PID in Padula et al [10], a neural in Ortolani et al [11], a fuzzy logic in Shieh et al [12], a model predictive control in Sawaguchi et al [13] and Chang et al [14], but in these contributions the control of the DoA is not fully automatic. More concretely, the administration of the hypnotic is made automatically, but the administration of the analgesic is manually made by a clinician.…”

Nonlinear dynamics of the patient’s response to drug effect during general anesthesia

“…The results reported in this study suggest that the control performance has been shown to be more robust by 12% during induction phase as compared to the study in [4], where only Propofol was administered. The disturbance were not the same in both studies, so a performance comparison during maintenance is not possible.…”

“…The inherent capability of MPC to outperform other control strategies has been already shown in [19,4] by means of simulations evaluating both the performance and the robustness of the closed loop system. In the general MPC scheme represented in Fig.…”

“…However, the proposed model based predictive control strategy has an intrinsic robustness to model variations and stability has been shown to be guaranteed for significant gain variations in [4], i.e. variations between 1:6 Ã k-0:5 Ã k, with k the gain of the nominal patient model.…”

“…Due to the fact that PID controllers cannot anticipate to the response of the patient and have limited robustness, stability problems may arise. Therefore, strategies using model based predictive control (MPC) [4] and Bayesian-based control algorithms [5,6] have been suggested and successfully applied in practice. The current bottleneck is the specific characterization of the multiple drug input and its synergistic effect upon the depth of anesthesia for personalized healthcare.…”