| Summary: | 博士 === 國立陽明大學 === 生物醫學影像暨放射科學系 === 100 === Exploring the interregional anatomical connections between cortical regions can reveal how the human brain is structurally organized into complex networks. Advanced diffusion magnetic resonance image (MRI) is the unique technique that can probe the white matter (WM) orientation with direction-dependent diffusivity of water molecules in vivo, and the further diffusion-based neural tractography can shows the WM pathway that have been widely used to reconstruct the fiber trajectories. By linking distinct cortical regions directly with fiber tracts, it is possible to map the whole brain WM connections. Moreover, with structural MRI, the covariance of gray matter (GM) morphology implies important structural or functional connectivity in the human brain, which has been used to construct the interregional connections indirectly. These techniques provide the opportunity to reveal the structural architecture of human brain in large-scale (known as human connectome).
With modern mathematical concept of graph theoretic analysis, the structural large-scale network of human brain has been demonstrated as a “small-world” topology (i.e., high local clustering and short path lengths linking the nodes), suggesting an optimal balance between structurally segregated and integrative organization. In this dissertation, first, we will summary recent methodological and application studies utilizing graph theoretic approaches on brain structural networks revealed by structural MRI and diffusion MRI. This work not only help us understand how the healthy human brain is structurally organized, but also provide a novel insight into the biological mechanisms of brain disorders. Second, this dissertation aims to construct the WM anatomical network for understanding of the information transmission among brain regions, and apply it as new approach for the study of human brain in Alzheimer’s disease (AD) followed by graph theoretical analysis.
Evidence from recent AD-related studies has shown that the impaired cognitive functions are accompanied by the disrupted regional connectivity with WM degeneration. Previous studies have demonstrated that the altered large-scale topological patterns in brain functional and structural networks in AD. However, the influence of WM dysconnection on the coordination of WM anatomical network in AD is still unknown. In this dissertation, we constructed the whole brain WM weighted networks using diffusion tensor image (DTI) tractography to investigate the AD-related alterations of topological architecture of WM networks with graph theoretical approaches. We found that the WM network of both AD patients and controls had a small-world topology. Furthermore, we showed that WM network contained highly connected hub regions that were predominately located in the precuneus, cingulate cortex, dorsolateral prefrontal cortex, and several occipital regions, which was also consistent with the previous diffusion-MRI studies. Compared with control networks, AD networks revealed the increased characteristic path length, normalized characteristic path length and decreased global efficiency, implying the imbalanced architecture that the WM anatomical networks are more in favor of the regular lattice for AD patients. We further found the reduced nodal efficiency in the frontal and temporal cortex regions of AD network. Moreover, the alterations of various network properties were significantly correlated with the cognitive and memory performances. These results could have implications for the understanding of how the abnormalities of structural connectivity in AD underlie behavioral deficits in the patients.
In summary, our work provides a network-level representation from regional to global view for the investigations of the human brain. The topology of brain network may be a major potential impact for our understanding of brain alterations either damage or reorganization in diseases. Knowing the architecture of brain networks in disease may help us to discover new biomarkers or evaluate the treatment in the future.
|
|---|
