Premium
P1‐234: IMPAIRED SEGREGATION OF THE VENTRAL DEFAULT MODE NETWORK IN ALZHEIMER'S DISEASE
Author(s) -
Jones David Thomas,
Gunter Jeffrey,
Wiste Heather,
GraffRadford Jonathan,
Machulda Mary M.,
Vemuri Prashanthi,
Boeve Bradley F.,
Knopman David S.,
Petersen Ronald,
Jack Clifford
Publication year - 2014
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2014.05.473
Subject(s) - default mode network , neuroscience , temporal lobe , atrophy , hippocampus , neuroimaging , correlation , computer science , artificial intelligence , psychology , functional connectivity , pattern recognition (psychology) , medicine , mathematics , geometry , epilepsy
for each subject the correlation between grey matter and functional eigenvector centrality across the nodes of a network and compared these relationships between the groups. Results: Group differences were similar for grey matter and functional graphs, where the average eigenvector centrality was lowest in the AD group and highest in the healthy control groups (structural p permuted 1⁄4 0.001; functional p permuted 1⁄4 0.04). MCI also showed significantly lower functional eigenvector centrality than controls (p permuted 1⁄4 0.03) and higher values than AD patients but this difference was not significant (p permuted 1⁄4 0.28). AD patients showed the lowest grey matter eigenvector centrality values in frontal and medial temporal structures (resp. p permuted 1⁄4 0.002; p permuted 1⁄4 0.008), and the lowest functional eigenvector centrality values in the occipital cortex including left superior occipital gyrus, calcarine sulcus and cuneus (p permuted 1⁄4 0.04). All individuals showed a moderately strong significant correlation between structural and functional eigenvector centrality that was strongest in the healthy controls (r 1⁄4 .30 6 0.08) intermediate in AD patients (r 1⁄4 .25 6 .1) and weakest in MCI (r 1⁄4 .23 6 .08). Conclusions: AD affacts global eigenvector centrality values in grey matter and functional graphs in a similar way. These findings could not be explained by differences in brain volume, gender or age. Because we used single-subject measurements of grey matter and functional graphs we were able to show a moderately strong relationship between grey matter and functional EC values for the first time. Our findings are in linewith a previous study that investigated group-derived grey matter graphs and reported a relationship with functional connectivity of similar strength (Alexander-Bloch et al., 2013). The relationship we found was weaker in AD and MCI, which might by due to the fact that AD targeted occipital areas in functional graphs and frontal and temporal structures in greymatter graphs. Our results support that a relationship exists between functional and grey matter graphs, and that this is modulated by disease.