21 Brain Morphological and Functional Networks …
331
matter using these images [2]. Anatomical MRIs are simple to acquire and are not
limited by artifacts to the same degree as other MRI-based techniques.
To bridge the above experimental limitations, and provide a new insight into
macro-scale brain connectivity and its advantages, my focus in this review is on
corticocortical networks extracted from anatomical and functional MRI, namely
morphological and functional networks. Corticocortical morphological networks are
extracted using T1-weighted anatomical MRI, which is a non-invasive assessment of
the brain’s structures at a sub-millimeter spatial resolution. Likewise, corticocortical
functional networks are extracted using functional MRI (fMRI), which records brain
activity via Blood-Oxygen-Level-Dependent (BOLD) signal as a proxy of neural
activity at the whole brain level. For the purpose of this article, I will review evidence
of (i) the corticocortical connections that are mediated by similarities in the cortical
morphology (i.e., cytoarchitecture) (ii) the relationship between functionally relevant
regional co-activity and underlying cytoarchitecture that may induce synchronized
plastic changes among related brain areas (i.e., activity-dependent plasticity) and (iii)
implications of this relationship for neurodegenerative syndromes. From the graph
theory perspective, my focus is (i) on the modular organisation of brain functional
and anatomical (morphological) networks, (ii) the time-varying modular topology of
functional interactions and (iii) on describing the potential of modular interactions
to inform theoretical and practical approaches to problems in neurodegenerative
syndromes.
21.2 Graph Theory and the Brain
One of the mathematical frameworks for studying the human brain structural (and
functional) organisation is graph theory. The brain network (graph) is modeled as a
set of nodes and edges. Nodes and edges are elementary building blocks of networks
and the definition of a node or an edge is of critical importance to the resulting
brain network models [15, 83]. The arrangement of nodes and edges defines the
organisation of the network, whose topology is quantified using statistical tools of
graph theory. Another major property of brain networks is the discovery that they are
modular by their topological organisation—they can be decomposed into groups of
nodes that are more densely connected to each other than with the rest of the network.
In what follows, I will describe in greater details these critical brain network elements
and their topological properties, with reference to the two brain networks in focus.
21.2.1 Brain Network Node
A challenging question in the field of large-scale MRI-based brain network analysis is: how to define meaningful nodes for a brain network? The solutions range
from defining nodes using the native resolution of the MRI technique (i.e., voxel-
331
matter using these images [2]. Anatomical MRIs are simple to acquire and are not
limited by artifacts to the same degree as other MRI-based techniques.
To bridge the above experimental limitations, and provide a new insight into
macro-scale brain connectivity and its advantages, my focus in this review is on
corticocortical networks extracted from anatomical and functional MRI, namely
morphological and functional networks. Corticocortical morphological networks are
extracted using T1-weighted anatomical MRI, which is a non-invasive assessment of
the brain’s structures at a sub-millimeter spatial resolution. Likewise, corticocortical
functional networks are extracted using functional MRI (fMRI), which records brain
activity via Blood-Oxygen-Level-Dependent (BOLD) signal as a proxy of neural
activity at the whole brain level. For the purpose of this article, I will review evidence
of (i) the corticocortical connections that are mediated by similarities in the cortical
morphology (i.e., cytoarchitecture) (ii) the relationship between functionally relevant
regional co-activity and underlying cytoarchitecture that may induce synchronized
plastic changes among related brain areas (i.e., activity-dependent plasticity) and (iii)
implications of this relationship for neurodegenerative syndromes. From the graph
theory perspective, my focus is (i) on the modular organisation of brain functional
and anatomical (morphological) networks, (ii) the time-varying modular topology of
functional interactions and (iii) on describing the potential of modular interactions
to inform theoretical and practical approaches to problems in neurodegenerative
syndromes.
21.2 Graph Theory and the Brain
One of the mathematical frameworks for studying the human brain structural (and
functional) organisation is graph theory. The brain network (graph) is modeled as a
set of nodes and edges. Nodes and edges are elementary building blocks of networks
and the definition of a node or an edge is of critical importance to the resulting
brain network models [15, 83]. The arrangement of nodes and edges defines the
organisation of the network, whose topology is quantified using statistical tools of
graph theory. Another major property of brain networks is the discovery that they are
modular by their topological organisation—they can be decomposed into groups of
nodes that are more densely connected to each other than with the rest of the network.
In what follows, I will describe in greater details these critical brain network elements
and their topological properties, with reference to the two brain networks in focus.
21.2.1 Brain Network Node
A challenging question in the field of large-scale MRI-based brain network analysis is: how to define meaningful nodes for a brain network? The solutions range
from defining nodes using the native resolution of the MRI technique (i.e., voxel-