Meeting Banner
Abstract #2258

Graph Analysis of Resting-State ASL Data Reveals Nonlinear Correlations Among CBF and Network Metrics

Xiaoyun Liang1, Alan Connelly1, 2, Fernando Calamante1, 2

1Brain Research Institute, Florey Institute of Neuroscience and Mental Health, Heidelberg, VIC, Australia; 2Department of Medicine, Austin Health and Northern Health, University of Melbourne, Melbourne, VIC, Australia

In this study, investigations on small-world network properties of ASL perfusion data have been conducted. The small-world network properties of ASL data are consistent with previous findings from BOLD data. Interestingly, the outcomes on the relationships between 4 specific network metrics and region-wise CBF demonstrate that consistent nonlinear patterns exist across normal subjects, which is well in line with the previous finding that hub regions tend to have higher values for degree, vulnerability and centrality, but lower values for characteristic path length, along with higher metabolic energy consumption (and CBF). This should not only provide useful information to further our understanding of the metabolic energy consumptions with which the brain can maintain normal cognition with the lowest cost, but also have applications for clinical studies on brain disorders.