A Neurocomputational Model Of The DG-CA3 Region Associated With Alzheimer's Disease And A Study Of The Firing Rhythm

Alzheimer’s disease（AD）is an irreversible neurodegenerative disease,which is clinically manifested by cognitive impairment such as learning disability and memory decline.Previous studies have shown that synaptic loss is an important pathological feature of AD and relevant experiments have also confirmed that the hippocampus is one of the first brain regions affected by AD.In view of this,this study combined neurodynamics and power spectrum analysis to study the typical biological neural network models and artificial neural network models in the hippocampus.To better understand the importance of synaptic loss on the pathogenesis of AD,reducing synaptic coupling parameters to characterize the synaptic loss of AD,the effect of synaptic loss on the firing rhythm of neurons in the hippocampus is discussed from the perspective of neural computation.The main contents and conclusions of this work are as follows:1.The DG-CA3 biological network model associated with Alzheimer’s disease and its dynamics have been studied.Considering that physiological experiments have shown that the Shank3 protein,one of the Shank protein subtypes closely related to glutamate synapses,is significantly reduced in the brains of AD patients.Therefore,the AMPA/NMDA-type synaptic coupling parameter from mossy cell to pyramidal neuron as well as the AMPA-type synaptic coupling parameter from pyramidal neuron to O-LM cell in the DG-CA3 biological network model were used to simulate the pathological feature of AD,which is glutamate synaptic loss caused by the decrease of Shank3 protein.Further,the correlation between synaptic loss and abnormal firing rhythm of neuron in AD is investigated by numerical simulation of firing rate and power spectrum.The result shows that synaptic loss of glutamate resulted in reduced excitability of neuron and reduced alpha rhythm,which was manifested by decreased firing rate of neurons and decreased relative alpha band power.These findings are consistent with the EEG of AD,indicating that the brain alpha rhythm slows down in patients with AD.The results of this study help to emphasize the important role of glutamate synaptic loss in the pathological process of AD.2.The DG-CA3 artificial network model associated with Alzheimer’s disease and its dynamics have been studied.Firstly,based on the physiological and anatomical structure of hippocampus and Hopefield artificial neural network,an artificial neural network model of DG-CA3 region is constructed.The synaptic connections between neurons are modified by the improved Oja learning rule.The synaptic coupling parameter from DG to CA3a and CA3b to CA3c is adjusted to simulate the synaptic loss in AD pathological state.The effect of synaptic loss on the firing rhythm of neuron is studied by numerical simulation.The result shows that with the decrease of synaptic coupling parameters,the excitability of CA3a/CA3c neurons decreased and the alpha rhythm slowed down.This result does not depend on the synaptic learning rate of Oja learning rule,and with the increase of learning rate,the relative power of alpha band decreases more rapidly with the decrease of synaptic coupling parameters.The results of this study further reveal the correlation between synaptic loss and abnormal firing rhythm of neuron in AD from the perspective of artificial neural network,highlighting the importance of synaptic loss in the pathological process of AD.