Dynamics Process Of Biological Neurons Action Potential

Neuroinformatics is a field which develops on the basis of computer science,mathematics,bio-electromagnetic technology and other interdisciplines.The study on neuroinformatics can promote the development of artificial intelligence,neuromorphic computing,channelopathies treatment and so on.Neuron is the basic functional units and structures of the nervous system.And action potential is used to transmit information between neurons.In this thesis,the single Hodgkin-Huxley(HH)model is taken as the study object.Non-linear dynamics theory and simulations are used to analyze HH model.And finally the dynamics of HH model is implemented by hardware.The main contents are listed below.(1)Selection of neuron models and basic theory of non-linear dynamics.The structure of neurons,the opening and shutting mechanism of ion channel and the basic electrophysiological activities of neurons are explained in this part.HH model which is most suitable for the study is selected as the object by analyzing development history of several common neuron models and their methods of describing neuron electrical activity.The parameters in HH model have clear physiological meanings and its structure conforms to biological reality which makes it suitable for the study of single neuron.And the basic theory used in the analysis is introduced.(2)Dynamic process analysis and numerical simulation of neuron action potential.The single HH model is used as the object of the research.Several parameters with definite physiological significance are taken as bifurcation parameters: stimulation current,sodium channel reversal potential,potassium channel reversal potential and leakage channel reversal potential.The dynamic process of action potential of neurons is studied by the method of numerical simulation and theoretical analysis of non-linear dynamic bifurcation.The results show that the change of stimulation current will cause the change of system stability,and the action potential of neurons will show different response under different stimulus intensity.When three ion channel reversal potentials are used as bifurcation parameters,action potentials are also periodically released by neurons,which may be the cause of channelopathies.(3)Hardware implementation of dynamic process of neuron action potential.The HH model is modeled and simulated under the mixed compiling environment of Matlab,DSP Builder and Quartus II.The action potentials of HH neuron model are implemented by using Field Programmable Gate Array(FPGA).The DSP Builder model simulation results of bifurcation parameters are compared with the results of hardware implementation,and both of the results are consistent.The results show that the hardware experimental platform can be competent to the study of the dynamic process of neuronal action potential.