Image Processing Reference
In-Depth Information
recently, HARDI glyphs have been combined dynamically with DTI glyphs and
fiber tracts based on local data characteristics [ 73 ].
21.4.3 Functional MRI
Blood-oxygen-level dependence, or BOLD, is a special type of MRI that is able to
measure increased levels of blood oxygenation [ 67 ]. Due to requiring more glucose
from the bloodstream, active neurons cause higher blood oxygenation in nearby
veins. Based on this principle, functional MRI, or fMRI, uses BOLD to image time-
dependent 3-D neural activity in the brain [ 68 ].
fMRI can also be used to derive functional or effective connectivity in the
brain. Functional connectivity is determined by calculating the temporal correla-
tions between the fMRI signals originating from different parts of the brain [ 32 ].
This is done either whilst the subject performs a specific task, in order to assess
how the brain network is applied during that task, or during resting state, in order to
derive the baseline functional brain network. Connectivity data can be determined
between a specific seed region or voxel and one or more other regions or voxels, or
exhaustively between all regions or voxels in the brain.
Effective connectivity, defined as the causal influence one neuronal system exerts
over another, is dependent on a model of the connectivity between the participating
regions. For example, the signal at one position could be expressed as the weighted
sum of the signals elsewhere [ 32 ]. If the model is invalid, the effective connectivity
derived from fMRI is also invalid.
Visualization of fMRI-derived connectivity information is quite varied, often
combining techniques from scientific and information visualization. Scatter plots
have been used to plot correlation strength over distance, dendrograms and multi-
dimensional scaling to represent correlations between regions in 2D [ 80 ], matrix
bitmaps to represent region-wise correlation matrices [ 28 ], 2-D and 3-D
(pseudo-) anatomical node-link diagrams to show the derived brain networks [ 107 ],
and coupled-view visual analysis techniques to explore resting state fMRI data [ 99 ].
When connectivity is determined between all pairs of voxels in the cortex, visual-
ization and knowledge extraction pose perceptual and computational challenges that
have not yet been fully explored.
21.5 Mesoscale Connectivity
Light microscopy was the first modality that allowed for imaging of single neuronal
cells. While the resolution of a light microscope is not sufficient to resolve synapses,
it allows for the identification of major cell parts, such as dendrites, somas, axons, and
also boutons as possible locations for synaptic connections. Imaging whole neuronal
cells and analyzing their geometry enables neuroanatomists to identify different types
 
Search WWH ::




Custom Search