Human brain networks in cognitive decline: a graph theoretical analysis of cortical connectivity from EEG data

The aim of this study was to investigate the neuronal network characteristics in physiological and pathological brain aging. A database of 378 participants divided in three groups was analyzed: Alzheimer's disease (AD), mild cognitive impairment (MCI), and normal elderly (Nold) subjects. Through EEG recordings, cortical sources were evaluated by sLORETA software, while graph theory parameters (Characteristic Path Length \u3bb, Clustering coefficient \u3b3, and small-world network \u3c3) were computed to the undirected and weighted networks, obtained by the lagged linear coherence evaluated by eLORETA software. EEG cortical sources from spectral analysis showed significant differences in delta, theta, and alpha 1 bands. Furthermore, the analysis of eLORETA cortical connectivity suggested that for the normalized Characteristic Path Length (\u3bb) the pattern differences between normal cognition and dementia were observed in the theta band (MCI subjects are find similar to healthy subjects), while for the normalized Clustering coefficient (\u3b3) a significant increment was found for AD group in delta, theta, and alpha 1 bands; finally, the small world (\u3c3) parameter presented a significant interaction between AD and MCI groups showing a theta increase in MCI. The fact that AD patients respect the MCI subjects were significantly impaired in theta but not in alpha bands connectivity are in line with the hypothesis of an intermediate status of MCI between normal condition and overt dementia