Probing Multi-Scale Circuitry And Neural Dynamics Associated With Acupuncture

The acupuncture effect is closely related to the nervous system.When being applied to the acupoint,the acupuncture stimulation of different frequencies can induce diverse multi-scale responses from spinal neurons to the brain within the neural circuits.And these multiscale response characteristics are closely related to intrinsic property of neurons and network,so that establishing the relationship between acupuncture manipulation,intrinsic properties of neuron,network structure and multi-scale neural response is the key to quantifying acupuncture manipulation and revealing functions of the neural circuit.Therefore,this thesis designed acupuncture experiments to probe multi-scale response of neural circuits evoked by acupuncture,constructed acupuncturing neuron and neural network models to link the structure of the neural circuit and multi-scale activity with functions,and further explored action mechanism and quantification of acupuncture manipulation.When being applied to the acupoint,the acupuncture manipulation can evoke a complex cerebral response with high dimensionality.Therefore,this thesis proposed to explore macro action mechanism of acupuncture from low-dimensional dynamics of cerebral response.Based on electroencephalogram(EEG)signals recorded from the experiment of acupuncturing at ST36(Zusanli)acupoint,we quantified the rhythmic activity of the brain by power spectrum method,and found the resonance is a way by which brain responds to periodic acupuncture stimulation.Moreover,we extracted the low dimensional dynamics of electrical activity in brain by utilizing dimensional reduction techniques,and found the variance of low-dimensional characteristics of cerebral response to the acupuncture under different acupuncture manipulations.In addition,the acupuncture-dependent component was further decomposed from brain activity,which was used to characterize and quantify the modulation of cerebral response to acupuncture.As the primary sensory neuron,the intrinsic properties of dorsal root neurons are closely related to the response of neural circuits to acupuncture.In order to probe the microscopic action mechanism of acupuncture on neural circuitry,this thesis constructed the functional model of dorsal root neurons under acupuncture manipulation.First of all,we designed an animal acupuncture experiment and collected the electrical activity of dorsal root neurons.The frequency characteristics of neuronal response in dorsal root neurons were analyzed by using the spike train obtained from experiments.It was found that different types of neurons responded differently to acupuncture stimulation.Moreover,functional model describing the response of dorsal root neurons to acupuncture was data-driven constructed based on the recurrent neural network(RNN)with force-learning.The systematical analysis indicated that the multiple time and wide spatial scale integration of input signal by single neuron is the key to realize different types of response.In order to establish the functional relationship between network structure and neural circuit and depict its influences on neural response,we constructed the computational model of neuronal network,and obtained the relationship between network structure and resonance,and found that stochastic resonance is an important way by which the neuronal network responds to periodic stimulation.Particularly,based on the anatomical structure of mammalian cortex,a heterogeneous neuronal network model with electrical and chemical synapses was established.Through numerical simulation and theoretical analysis,it was demonstrated that the time delay can induce the phenomenon of multiple stochastic resonances in neuronal network.Both electrical synapses and chemical synapses can enhance the resonance response.The interaction of excitation-inhibition is able to alter the intrinsic frequency of the neuronal network and thus affect the selectivity of neuronal network for input signals.According to the intrinsic property and resonance mechanism of neuronal network obtained above,we inspected the spontaneous electrical activity of neural population in rats,and further obtained the relationship between intrinsic property and network activity.In particular,the network spontaneously oscillated with multiple intrinsic frequencies.Moreover,we analyzed the firing characteristic of neurons and found the fluctuation scaling among neuronal spike trains.This scale invariance on the time scale is one of the important causes reducing dimensionality in neuronal network.Overall,this thesis provides important computational tools for probing both microand macro-scale neural circuitry and its dynamics,which is crucial for understanding the mechanism of central nervous systems.The study is of great theoretical and practical significance for scientifically quantifying acupuncture manipulation,explaining the mechanism of acupuncture and improving its effect.