Design of adaptive neuro sliding mode controller for anesthesia drug delivery based on biogeography based optimization

Abstract

Monitoring depth of anesthesia (DOA) is a significant point in general anesthesia (GA). It can be obtained from the assessment of the drug dose carefully and preciously. As a benefit of drug delivery automation, closed-loop method will present several advantages. It may prevent excessive dose amount or less needed dose and the controlled feedback system can decrease the cost of the healthcare by reducing the patient recovery period. This paper addresses the use of adaptive sliding mode controllers (ASMC) for calculating the depth of anesthesia by administrating a dose of propofol drug and measure patient state according to the monitoring device the bispectral index (BIS). In this study, we suggest a simple nonlinear control strategy consists of an ASMC combined with single neuron self-tune neural controller. It is used for maintaining DOA and reducing the effect of the nonlinear element. The adaptive controller uses the BIS value measured as a reference tracking value and propofol dose rate as a control signal. The parameters of the controller are tuned using a procedure based on the biogeography-based optimization (BBO) algorithm. The results indicate that including adaptive parts of the controller and tune its gains needed by BBO algorithm may enable optimal and stable performance for controller for all patients. It also provides fast reach to the induction phase and stay in a stable value in maintenance phase, which reflects the efficient response of the suggested controller if it compared to other nonlinear controllers. It is also justified by the results obtained that the suggested controller gives a very good response.

Physiological Closed-Loop Control (PCLC) Systems: Review of a Modern Frontier in Automation

“…As mentioned previously, there are four different types of controllers that are commonly used in the closed-loop control of anesthesia: proportional-integral-derivative, model predictive, adaptive, and fuzzy-logic controllers. Besides these controllers, strategies such as observer-controllers [213], robust and robust deadbeat controllers [220], [263], [328], nonlinear H-infinity controllers [329], non-overshooting tracking controller [330], sliding mode controllers [331], [332], and other types of nonlinear controllers [333] have been studied for use in anesthesia. Some of these studies have employed performance metrics such as settling-time, overshoot, undershoot, lowest observed BIS value (BIS-NADIR), performance error (PE), median performance error (MDPE), median absolute performance error (MDAPE), total variation (TV), integrated absolute error (IAE), and WOBBLE (an index of response variations over time) to compare their results and controller effectiveness to other control strategies in attempts to demonstrate advantages over other schemes and find the most suitable class of controllers for anesthesia [272], [334]- [336].…”

An Efficient K-Means Method Based on Centroid Handling for the Similarity Estimation

“…Standard data mining techniques have been used widely in the distance estimation tasks [13]. The concrete structure of any data mining exploration relies on the appropriate data representation in terms of the data scenic approach for the collaborative approach of the data centric approach in terms of data exploration and data recognition [14][15][16]. It also explores the mechanism for the adaptation of the correct methodology in terms of data mining methodology in terms of suitability and use.…”