Item

Abstract

The typical hallmark of electroencephalogram (EEG) in Alzheimer’s disease (AD) is a slowing of rhythms and perturbations in synchrony. However, the mechanism of AD electrophysiological abnormalities is still ambiguous. Synapse deficiency has been considered as an evident neuropathological change in AD that is closely associated with cognitive decline. The main purpose of this work is to explore how synapse deficiency in AD affects these electrophysiological features using neural computational techniques. First, based on the Diffusion Tensor Imaging data, a connectivity matrix of a structural brain network is constructed by means of a pipeline toolbox called PANDA. Using this data-driven connectivity matrix, a cortical network model with 90 cortical areas is then be built in which each cortical area is modeled by a neuron mass model. Subsequently, by reducing the synaptic strength parameter to mimic synapse deficiency in AD, our results show that the synapse deficiency does not only cause a leftward shift of the dominant frequency, but also induces a decrease in the alpha rhythm and an increase in the theta rhythm. Further, the influence of synapse deficiency on phase synchrony is investigated by the phase lag index (PLI). When the synaptic strength parameter is reduced, the alpha-band PLI decreases and theta-band PLI increases. Moreover, a statistical analysis of the differences between the simulated AD and healthy control (HC) in terms of synchronization and rhythms is performed. The results demonstrate that there are significant differences between simulated AD and HC groups. All the above simulation results are consistent with the EEG changes of AD in the physiological experiments. Finally, a strong statistical correlation between PLI and relative power is revealed using Pearson’s correlation analysis. This study reveals a close relationship between synapse deficiency and electrophysiological abnormalities in AD, which may provide new insight for the early diagnosis of AD.