Time-frequency and Network Analysis of Hippocampus in APP/PS1 Double Transgenic Mice with Dementia of Different Degrees

Abstract

Hippocampal atrophy and neuron loss are common symptoms of Alzheimer's disease (AD). The hippocampal region is well known for producing oscillations at different frequency bands due to the neuronal network architecture. However, the mechanism of Ripple high frequency variation in hippocampal region with the course of AD disease has not been correctly assessed. We proposed time-frequency analysis using wavelet transform and constructing Granger causality network to analyze the characteristics of Hippocampal sharp wave-ripple (SPW-R) complexes in APP/PS1 mice at different cognitive levels. We use wavelet transform to overcome the shortcoming that the traditional Short Time Fourier Transform cannot deal with the unsteady signal frequency, and construct the Granger causality network to verify our results. By analyzing ripple frequency band energy changes and directional transfer function matrix in hippocampal CA1 region of mice with different cognitive levels, we found that the loss of ripple high frequency energy and decreased network connectivity in hippocampal CA1 region of APP/PS1 mice were correlated with the degree of memory loss. We believe that from mild dementia to severe dementia. The decreased cell activity in APP/PS1 mouse CA1 region leads to changes in Ripple high-frequency time-frequency energy and network connectivity for theoretical reasons. Our results provide support for assessing cognitive loss in APP/PS1 mice from the perspective of Ripple high frequency in hippocampus CA1 region.