Parameter Estimation And Control Of Epileptic Seizures Neuronal System

Epilepsy is one of common chronic diseases of the nervous system, almost fifty millionpeople survive with epilepsy in the world. More than90%of patients with epilepsy live indeveloping countries. The main characteristics of seizure are sudden and repetitive discharging,as soon as the seizure occurs, the spasm, lossing of consciousness and other terrible symptomsarise. At the same time, the risk of premature death of seizure is two or three times than that ofthe normal people, leading to huge burden to patients and society. All the time, the majortreatment for epilepsy is drug therapy and surgery, but for the remain30%of the patients, noeffective methods are useful momentarily. Thus, it is urgent to study the mechanism of epilepsy,so that the abnormal epileptic discharge behavior can be controlled.Physiological experiments are difficult to realize, therefore the neuron model which hassubstituted physiology experiments has become the effective method in exploring the mechanism.Hodgkin-Huxley (HH) model is the most fundamental neuron model which describes how theneuron works. In this paper, in order to study the most basic mechanism of epileptic, we extendthe HH model to study the dynamical characteristics of the model, and find the key parameterswhich are able to influence the discharging characteristics, such as sodium concentration,potassium concentration. Stabilizing the balance of ion concentration plays an importance role inthe normal activities of neurons. In this work, we attempt to use these parameters as the feedbacksignal to design the control method to eliminate epileptic discharge behavior.However, in the “real world”, due to the abnormal stimulations, some neuronal parameterscan be changed, and then the physiological activities of the neuronal system become abnormal,leading to various neurological diseases. The neuronal model is characterized by the nonlinearity,so that some model parameters of these neurons can not be measured directly. In addition,because of the noise which exists in the neuronal system, the neuronal state can not be obtainedexactly. Thus, we propose an appropriate nonlinear parameter estimation method to estimate theneuronal state behaviors and the unpredictable parameters. The proposed approach is helpful tostudy the underlying mechanism of the seizures and treatment methods. Simulation resultsdemonstrate that the validity of the proposed methods. Accurate parameter estimation contributesto build a more real neuron model, and reveal the working mechanisms of neurons well.Suppression of epilepsy seizure is the basic problem in epilepsy. Currently, due to theunclear epilepsy pathogenesis, the treatment can only alleviate illness but not cure. Therefore, itis necessary to study the underlying mechanism of neuronal pathological dynamics and feasiblecontrol methods which can realize the inhibition of neuronal abnormal discharging. In this paper, two methods which contains of the open-loop and closed-loop control can effectively suppressthe neuronal abnormal discharging behavior. The reasonable control strategy is helpful tounderstand the information transmission and processing mechanism of the brain, thus themethods can be used as a potential neurological diseases treatment.