Model Analysis And Parameter Identification Of Thalamic-Cortical Coupling Neural Mass

Based on the average field modeling idea,the neural mass model(NMM)reflects the average behavior of the whole neuronal population in the neural networks with the lumped state variables,which is an important tool for studying the physiological mechanism of neural oscillation.Simulation signals with different properties can be obtained by adjusting model parameters,which have physiological significance.Studying the influence of these parameters on the model outputs is helpful to understand the generation mechanism of EEG signals.In this paper,a thalamic-cortical coupled NMM was established to analyze the effect of the synaptic connection parameters on the outputs of the model.Then,the unscented Kalman filter and genetic algorithm were applied to the thalamus-cortical coupled NMM for its parameters identification,and the mechanism of mild cognitive impairment has been studied.Firstly,a modified thalamic NMM was established,in which the variation of two parameters will cause the decline of Alpha band of the output signal in the simulation analysis.Then,based on the thalamus-cortex coupled NMM modeling,analyzed the influence of synaptic connection parameters on the high frequency components of the output signals of cortical module output signals in the coupling model,and the results illustrate that the synaptic connection parameters between the retinal cell population and the thalamocortical relay cell population is proportionate to the high frequency components of the output signals.Secondly,genetic algorithm and UKF filter were used to identify the parameters of the thalamic-cortical coupling neural mass model.The simulation results showed that the estimation of model parameters using genetic algorithm and UKF filter had high accuracy.By identifying the model parameters,the EEG signals were connected with the model parameters,which provided a feasible method to study the generation mechanism of EEG signals.Finally,the UKF filter was used to identify the physiological parameters in the EEG data of patients with mild cognitive impairment.The results showed that the estimated value of the connection coefficient between the retinal cell population and the thalamic cortex relay cell population in MCI group was lower than that in the control group.Moreover,there were significant differences between MCI group and control group in the prefrontal,occipital,parietal,left temporal and right temporal regions.The results of this study have great significance for the study of the pathogenesis,diagnosis and treatment of neurological diseases such as mild cognitive impairment.