Sergio Islas Andrade

Robust control for propofol induced anesthesia based on Second-Order Sliding-Mode Control

The effect of anesthesia can be different in every patient. It is often the case that anesthesia has to be induced in emergency situations, where identification of the patients parameters cannot be done. It is preferable to have a controller insensitive to parameter variations, such as Second-Order Sliding-ModeControl (SOSMC). This kind of controller is also robust with respect to singular perturbations that may occur during surgery; therefore SOSMC is well suited to perform a closed loop propofol dosage. This in silico study tests the performance of an SOSMC for 4 different patients, to evaluate its insensibility to parameter variation (interpatient variability). SOSMC robustness with respect to perturbations is evaluated by a bleeding simulation, that changes the parameters of the pharmacokinetic model (intrapatient variability), and also by inducing a surgical stimuli.

High-Order Sliding-Mode control of blood glucose concentration via practical relative degree identification

This research introduces a novel concept of practical relative degree and presents a numerical method of practical relative degree identification. The concept efficacy is demonstrated by computer simulation of a High-Order Sliding-Mode controller, effectively stabilizing the blood glucose concentration for two well known models with different relative degrees.

Experimental glucose regulation with a High-Order Sliding-Mode Controller

Theoretically High-Order Sliding-Mode Controllers are well suited to perform closed loop glucose regulation because they are insensitive to parameter uncertainties and robust to unknown dynamics that may perturb the system. The implementation of the controller based on the concept of practical relative degree is presented. The controller was tested in Sprague-Dawley rats with steptozotocin induced diabetes. The tests demonstrated high efficacy and robustness of the controller.

High Order Sliding Mode Controller for blood glucose in type 1 diabetes, with relative degree fluctuations

Glucose is regulated by insulin injections in patients with diabetes type 1. High Order Sliding Mode Controllers are well suited to perform a closed loop control due to is a non linear control, robust with respect to parameter uncertainties, and model unaccounted dynamics. These features represents an advance to other works in this field, due to the system is non linear, variable in the time, and the parameter identification is expensive and invasive to the patient. The improving of the model of glucose-insulin regulatory system, considering additional dynamics, increases, arbitrarily, the model order. The theorem that ensures the robustness of HOSMC with respect to unaccounted dynamics is given. This theorem justify the implementation of a controller designed for a reduced model will be able to handle the dynamics of a extended model. In this paper the theorem is illustrated by the design of a third order sliding mode controller, based on BeM, and tested also in SoM that is one of the most complete models and have 24 differential equations and has relative degree five.

High-order sliding-mode control for anesthesia

Depth of anesthesia can be indirectly measured by Bispectral Index (BIS), therefore it is possible to administer propofol in a closed loop to maintain the optimal level of anesthesia while minimizing the dose to improve the postanesthesia recovery. High-Order Sliding-Mode control can be used to individualize drug dosing. In this study, the controller is tested with four in silico patients.