Complex Brain Networks: A Graph-Theoretical Analysis

241

subjects whereas the connections of PCC were reduced in AD patients. Ac-

cording to the authors, decline in memory functions may be attributed to this

effect and thus, the activity in PCC may be used to determine the level of AD

progression in patients providing a potential biomarker.

Local network connectivity changes in AD patients using graph analysis in

fMRI were searched in [39]. 18 mild AD patients and 21 healthy age-matched

control subjects were tested to find that the characteristic path length of AD

functional networks is closer to that of random networks whereas the clus-

tering coefficient values were similar to that of control subjects. The average

synchronization of the brain in both groups were similar, however, increased

synchronization related to frontal cortices and decreased synchronization at

the parietal and occipital regions were detected in Ad patients. The authors

conclude that long distance connectivity is affected and the randomization of

functional brain network structure is observed which probably is responsible

for the loss of global information integration in AD.

The default mode network (DMN) is a canonical large-scale brain network

consisting of distributed nodes that show increased and correlated BOLD

response during wakeful rest. The characteristics of brain networks in subjects

with AD is observed as the loss of small-world features and tendency to change

into a randomized and/or regular network topology; change in hub regions

especially in the DMN and medial temporal lobe [40].

9.7.2

Schizophrenia Connectome

Schizophrenia is a psychiatric disorder with heterogeneous symptoms among

affected individuals. The cause of this disease which generally starts at early

adulthood is not known. Schizophrenia is often characterized by delusions, hal-

lucinations, disorganized thinking and behavior and impairment of cognitive

functions. Schizophrenia is commonly considered as a disorder of connectiv-

ity between large-scale brain network regions and as with other neurological

disorders, this disease is associated with impairment of brain connectivity

[41].

Structural brain networks were found to be less strongly integrated in pa-

tients with schizophrenia in [42] with key frontal hubs having a reduced central

role. Also, the network connectivity of frontal and temporal areas was signif-

icantly reduced and the path length was substantially increased in of olfac-

tory, medial, and superior frontal regions, anterior cingulate, medial temporal

pole, and superior occipital regions preventing efficient transfer between these

regions.

In a review of brain networks in schizophrenia [43], reduced structural

connectivity is reported with white matter projections linking frontal, tempo-

ral and parietal regions being the most affected. The authors state that the

average path length is increased and the communication between segregated

regions of the brain is reduced as verified in various studies. Moreover, it is

reported that a main characteristic of schizophrenia may be the pathology