MEG Connectivity Analysis in Patients with Alzheimer’s Disease Using Cross Mutual Information and Spectral Coherence

Abstract

Alzheimer's disease (AD) is an irreversible brain disorder which represents the most common form of dementia in western countries. An early and accurate diagnosis of AD would enable to develop new strategies for managing the disease; however, nowadays there is no single test that can accurately predict the development of AD. In this sense, only a few studies have focused on the magnetoencephalographic (MEG) AD connectivity patterns. This study compares brain connectivity in terms of linear and nonlinear couplings by means of spectral coherence and cross mutual information function (CMIF), respectively. The variables defined from these functions provide statistically significant differences (p < 0.05) between AD patients and control subjects, especially the variables obtained from CMIF. The results suggest that AD is characterized by both decreases and increases of functional couplings in different frequency bands as well as by an increase in regularity, that is, more evident statistical deterministic relationships in AD patients' MEG connectivity. The significant differences obtained indicate that AD could disturb brain interactions causing abnormal brain connectivity and operation. Furthermore, the combination of coherence and CMIF features to perform a diagnostic test based on logistic regression improved the tests based on individual variables for its robustness.

On the early diagnosis of Alzheimer's Disease from multimodal signals: A survey

“…A generalized loss of functional interactions (i.e. decreased synchrony) has also been found [202,207,208]. People with MCI have also been investigated with MEG, verifying that their symptoms are somewhere between those of AD and controls [203,209,210].…”

Regional coherence evaluation in mild cognitive impairment and Alzheimer's disease based on adaptively extracted magnetoencephalogram rhythms

“…MEG recordings are composed of numerous channels and it may be helpful to group them into regions (Alonso et al, 2011;. In this study, the channels are grouped into Left Central (LC), Anterior (LA), Lateral (LL), and Posterior (LP); and Right Central (RC), Anterior (RA), Lateral (RL), and Posterior (RP) regions as shown in Fig.…”

“…It has been hypothesised that AD decreases c(f ) in α and β but there are no conclusive findings about the δ and θ bands (Jeong, 2004;Rossini et al, 2007). Furthermore, AD produces opposite changes in features computed from different spectral bands (Alonso et al, 2011;Stam et al, 2009;van Cappellen van Walsum et al, 2003). Thus, it is advisable to analyse different spectral bands separately (Stam, 2010).…”

The role of nonlinearity in computing graph-theoretical properties of resting-state functional magnetic resonance imaging brain networks

“…13 The equiquantal mutual information estimator has been successfully used in detection and quantification of nonlinearity in the dynamics of various complex systems, ranging from the Earth's atmosphere 14 to the human heart 15 and brain. [16][17][18][19] In neuroscience, functional magnetic resonance imaging (fMRI) is one of the prominent methods for the study of large-scale brain dynamics. Following the increase of interest in the study of complex network properties, a wealth of graph-theoretical studies utilizing resting-state fMRI data has been published in the last 5 yr. 20 Importantly, linear connectivity measures such as Pearson or partial linear correlation of the local activity time series are used to derive the connectivity matrix that is then transformed to the network representation.…”