An Optimized Type-2 Self-Organizing Fuzzy Logic Controller Applied in Anesthesia for Propofol Dosing to Regulate BIS

During general anesthesia, anesthesiologists who provide anesthetic dosage traditionally play a fundamental role to regulate bispectral index (BIS). However, in this article, an optimized type-2 self-organizing fuzzy logic controller (SOFLC) is designed for a target controlled infusion pump related...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on fuzzy systems Vol. 28; no. 6; pp. 1062 - 1072
Main Authors: Wei, Zi-Xiao, Doctor, Faiyaz, Liu, Yan-Xin, Fan, Shou-Zen, Shieh, Jiann-Shing
Format: Journal Article
Language:English
Published: New York IEEE 01.06.2020
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Anesthesia
Automatic control
Brain modeling
Computational modeling
Computer simulation
Control systems design
Controllers
Dosage
Fuzzy control
Fuzzy logic
Fuzzy sets
General anesthesia
Genetic algorithms
genetic programming (GP)
Optimization
Surgery
surrogate model (SM)
type-2 fuzzy controller
ISSN:1063-6706, 1941-0034
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:During general anesthesia, anesthesiologists who provide anesthetic dosage traditionally play a fundamental role to regulate bispectral index (BIS). However, in this article, an optimized type-2 self-organizing fuzzy logic controller (SOFLC) is designed for a target controlled infusion pump related to propofol dosing guided by BIS, to realize automatic control of general anesthesia. The type-2 SOFLC combines a type-2 fuzzy logic controller with a self-organizing mechanism to facilitate online training while able to contend with operational uncertainties. A novel data-driven surrogate model and genetic programming-based strategy is introduced for optimizing the type-2 SOFLC parameters offline to handle interpatient variability. A pharmacological model is built for simulation in which different optimization strategies are tested and compared. Simulation results are presented to demonstrate the applicability of our approach and show that the proposed optimization strategy can achieve better control performance in terms of steady-state error and robustness.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1063-6706
1941-0034
DOI:10.1109/TFUZZ.2020.2969384