Premium
P4‐119: GRAPH THEORETIC ANALYSIS OF STRUCTURAL CONNECTIVITY IN INDIVIDUALS WITH NORMAL COGNITION, MILD COGNITIVE IMPAIRMENT, AND DEMENTIA DUE TO ALZHEIMER'S DISEASE
Author(s) -
Soldan Anja,
McGlaughlin Alec,
Ruth David,
Jager Leah,
Phillips David J.
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.2014.05.1635
Subject(s) - dementia , cognitive impairment , cognition , disease , psychology , node (physics) , alzheimer's disease , neuroscience , audiology , medicine , structural engineering , engineering
Background: Graph theory is emerging as a new technique to study changes in brain connectivity and network organization that occur with Alzheimer’s disease (AD). Although prior studies have demonstrated ADrelated changes in structural networks, these changes were often small and sometimes inconsistent, likely due to differences in network construction. The current study represents the first systematic investigation of how methods of structural network construction affect network properties. A second goal was to investigate the order in which network properties become abnormal with increasing disease severity. Methods: We estimated the structural connectivity of brain networks based on correlation maps of cortical thickness obtained from magnetic resonance imaging. Network properties of four diagnostic groups were compared: 126 cognitively normal older adults who remained normal for at least 3 years, 103 individuals with Mild Cognitive Impairment (MCI) who retained MCI status for at least 3 years (stableMCI), 108 individuals withMCIwho developed to AD-dementia within 3 years (progressive MCI), and 105 individuals with AD-dementia.Results:Network properties most reliably differentiated the four groups when the edges of the graphs were weighted by the normalized product of the mean cortical thickness of the two regions. While other methods of network construction usually produced the same pattern of results, between-group differences were less reliable or absent. Of the metrics examined, characteristic path length, a measure of how well each node (or brain region) is connected to every other node, became abnormal early in the course of the disease. It was significantly greater in subjects with stable MCI compared to normal subjects and continued to increase as the disease progressed. Assortativity, which expresses the tendency of nodes with similar connectivity to connect with each other, appeared to become abnormal later and was significantly smaller in individuals with progressive MCI compared to those with stable MCI. Lastly, algebraic connectivity, a measure of global network connectivity, seemed to decrease early (from Normal to stable-MCI) and then stabilized. Conclusions: These results may suggest that different graphmetrics of structural brain networks change in an ordered fashion as AD pathology develops and may be useful for predicting symptom onset and tracking progression.