Study On The Electrical Model Of Reconstruction Of Cortical EEG Signal By Scalp EEG

In this paper,by creating an electrical model of brain electrical signal transmission in the head,to simulate the process of cerebral cortex brain electrical signal(Electrocorticography,ECoG)transmission from cerebral cortex through dura,skull,muscle,fat and skin to scalp brain electrical signal(Electroencephalography,EEG),On this basis,by establishing the inverse model of this process,the reconstruction of cortical EEG signal from scalp EEG signal is studied,and it is applied to the research of epilepsy EEG signal in order to improve the accuracy of epileptic seizure detection(or prediction)from the perspective of signal source.Firstly,the basic electrical model of ECoG-EEG is established to study the parameters of signal attenuation and delay in the process of EEG transmission from cerebral cortex to cerebral scalp.By establishing the mathematical model of the dielectric properties of biological tissue changing with frequency,the dielectric constant and conductivity of each layer of human brain tissue were calculated,and the resistance and capacitance values in the basic electrical model were determined.Secondly,considering the inconsistency of the conduction path when the cortical EEG in different parts of the head is scalp EEG,and the absorption or attenuation of EEG signals in different spatial directions of the head tissues,an ECoG-EEG model considering the effects of spatial ambiguity and anisotropy was established.Then,the ECoG and EEG data in the public database were used to simulate the electrical model,and the changes of the spectrum,potential amplitude and phase of EEG signals transmitted from the cerebral cortex to the scalp at frequencies of 100 Hz and 10 k Hz were observed.The simulation results show that when the frequency is 100 Hz,the amplitude of EEG signal from cortex to scalp is attenuated by 31.5 times,and the phase lag is 0.0791degrees;when the frequency is 10 k Hz,the amplitude of EEG signal from cortex to scalp is attenuated by 5.7 times,and the phase lag is 0.0791 degrees.Furthermore,based on the electrical model of ECoG-EEG,an inverse model(EEGECoG)is established to reconstruct the cortical EEG from the scalp EEG.Firstly,the ECoGEEG model considering spatial ambiguity and anisotropy is simplified.The load in the simplified model is divided into resistance and capacitance,and the transfer function and its inverse transfer function were obtained.Secondly,according to the inverse transfer function,the inverse model of cortical EEG conduction as scalp EEG is established.When the frequency is 100 Hz,the cortical EEG conduction with resistance is scalp EEG.The inverse model is composed of amplifying circuit,summation circuit and integrating circuit;The inverse model is composed of amplification circuit,summation circuit and differential circuit.Finally,the inverse model is simulated and analyzed.The results show that when the load is resistor at 100 Hz,the amplitude of the reconstructed ECoG signal is 31.9 times that of the scalp EEG signal,and the phase lag is 0.046 degrees;When the load is capacitance at 100 Hz,the amplitude of the reconstructed ECoG signal is 31.9 times that of the scalp EEG signal,and the phase is 0.003 degree ahead.