Limits...
The envirome and the connectome: exploring the structural noise in the human brain associated with socioeconomic deprivation.

Krishnadas R, Kim J, McLean J, Batty GD, McLean JS, Millar K, Packard CJ, Cavanagh J - Front Hum Neurosci (2013)

Bottom Line: For example, the human brain has been found to have a modular architecture i.e., regions within the network form communities (modules) with more connections between regions within the community compared to regions outside it.These results suggest that apparent structural difference in brain networks may be driven by differences in cortical thicknesses between groups.These findings provide some evidence of the relationship between socioeconomic deprivation and brain network topology.

View Article: PubMed Central - PubMed

Affiliation: Sackler Institute of Psychobiological Research, Institute of Health and Wellbeing, University of Glasgow, Gartnavel Royal Hospital Glasgow, UK.

ABSTRACT
Complex cognitive functions are widely recognized to be the result of a number of brain regions working together as large-scale networks. Recently, complex network analysis has been used to characterize various structural properties of the large-scale network organization of the brain. For example, the human brain has been found to have a modular architecture i.e., regions within the network form communities (modules) with more connections between regions within the community compared to regions outside it. The aim of this study was to examine the modular and overlapping modular architecture of the brain networks using complex network analysis. We also examined the association between neighborhood level deprivation and brain network structure-modularity and gray nodes. We compared network structure derived from anatomical MRI scans of 42 middle-aged neurologically healthy men from the least (LD) and the most deprived (MD) neighborhoods of Glasgow with their corresponding random networks. Cortical morphological covariance networks were constructed from the cortical thickness derived from the MRI scans of the brain. For a given modularity threshold, networks derived from the MD group showed similar number of modules compared to their corresponding random networks, while networks derived from the LD group had more modules compared to their corresponding random networks. The MD group also had fewer gray nodes-a measure of overlapping modular structure. These results suggest that apparent structural difference in brain networks may be driven by differences in cortical thicknesses between groups. This demonstrates a structural organization that is consistent with a system that is less robust and less efficient in information processing. These findings provide some evidence of the relationship between socioeconomic deprivation and brain network topology.

No MeSH data available.


Related in: MedlinePlus

Shows the pipeline of analysis, including the parcellation schemes—Desikan atlas and Destrieux atlas showing the sulcogyral parcellations and the Finegrain 200 and 1000 atlas as in Echtermeyer et al. (2011).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3824100&req=5

Figure 2: Shows the pipeline of analysis, including the parcellation schemes—Desikan atlas and Destrieux atlas showing the sulcogyral parcellations and the Finegrain 200 and 1000 atlas as in Echtermeyer et al. (2011).

Mentions: Cortical reconstruction was performed with the FreeSurfer image analysis suite, which is documented and freely available for download online (http://surfer.nmr.mgh.harvard.edu/). (Dale et al., 1999; Fischl et al., 1999; Fischl and Dale, 2000) Briefly, following skull-stripping and correction of inhomogeneity artifact, constrained region growing was used to create a unitary white matter volume for each hemisphere. The gray-matter/white-matter boundary for each cortical hemisphere was determined using tissue intensity and neighborhood constraints. The white matter surface was tessellated by assigning two triangles to the square face of each surface voxel. This process yielded approximately 160000 vertices per hemisphere. The white matter surfaces were deformed toward the gray matter/pial boundary, with a point to point correspondence at each vertex. Cortical thickness was computed as the distance between the white and the pial surfaces at each vertex. Cross-subject registration of hemispheric cortical surfaces was performed by projecting them onto the spherical representations. The maps produced are not restricted to the voxel resolution of the original images and are thus capable of detecting sub-millimeter differences between groups. The parcellations were obtained using the Desikan sulcogyral-based atlas, which follows the anatomical conventions of Duvernoy. The FS image-processing pipeline was visually inspected and corrected at critical points in order to avoid errors permeating through the subsequent analyses. Procedures for the measurement of cortical thickness have been validated against histological analysis and manual measurements. The Desikan Killiany atlas produces 68 parcellations based on gyri and sulci (Desikan et al., 2006). In addition to the Desikan Killiany atlas parcellation scheme, we also used fine-grained parcellation schemes based on anatomical sulcogyral boundaries including the Destrieux atlas, (148 parcellations) and fine-grained parcellation schemes (200, and 1000 parcellations) that did not follow anatomical conventions described in Echtermeyer et al. (Destrieux et al., 2010; Echtermeyer et al., 2011). The pipeline of the analysis and the parcellation are shown in Figure 2.


The envirome and the connectome: exploring the structural noise in the human brain associated with socioeconomic deprivation.

Krishnadas R, Kim J, McLean J, Batty GD, McLean JS, Millar K, Packard CJ, Cavanagh J - Front Hum Neurosci (2013)

Shows the pipeline of analysis, including the parcellation schemes—Desikan atlas and Destrieux atlas showing the sulcogyral parcellations and the Finegrain 200 and 1000 atlas as in Echtermeyer et al. (2011).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3824100&req=5

Figure 2: Shows the pipeline of analysis, including the parcellation schemes—Desikan atlas and Destrieux atlas showing the sulcogyral parcellations and the Finegrain 200 and 1000 atlas as in Echtermeyer et al. (2011).
Mentions: Cortical reconstruction was performed with the FreeSurfer image analysis suite, which is documented and freely available for download online (http://surfer.nmr.mgh.harvard.edu/). (Dale et al., 1999; Fischl et al., 1999; Fischl and Dale, 2000) Briefly, following skull-stripping and correction of inhomogeneity artifact, constrained region growing was used to create a unitary white matter volume for each hemisphere. The gray-matter/white-matter boundary for each cortical hemisphere was determined using tissue intensity and neighborhood constraints. The white matter surface was tessellated by assigning two triangles to the square face of each surface voxel. This process yielded approximately 160000 vertices per hemisphere. The white matter surfaces were deformed toward the gray matter/pial boundary, with a point to point correspondence at each vertex. Cortical thickness was computed as the distance between the white and the pial surfaces at each vertex. Cross-subject registration of hemispheric cortical surfaces was performed by projecting them onto the spherical representations. The maps produced are not restricted to the voxel resolution of the original images and are thus capable of detecting sub-millimeter differences between groups. The parcellations were obtained using the Desikan sulcogyral-based atlas, which follows the anatomical conventions of Duvernoy. The FS image-processing pipeline was visually inspected and corrected at critical points in order to avoid errors permeating through the subsequent analyses. Procedures for the measurement of cortical thickness have been validated against histological analysis and manual measurements. The Desikan Killiany atlas produces 68 parcellations based on gyri and sulci (Desikan et al., 2006). In addition to the Desikan Killiany atlas parcellation scheme, we also used fine-grained parcellation schemes based on anatomical sulcogyral boundaries including the Destrieux atlas, (148 parcellations) and fine-grained parcellation schemes (200, and 1000 parcellations) that did not follow anatomical conventions described in Echtermeyer et al. (Destrieux et al., 2010; Echtermeyer et al., 2011). The pipeline of the analysis and the parcellation are shown in Figure 2.

Bottom Line: For example, the human brain has been found to have a modular architecture i.e., regions within the network form communities (modules) with more connections between regions within the community compared to regions outside it.These results suggest that apparent structural difference in brain networks may be driven by differences in cortical thicknesses between groups.These findings provide some evidence of the relationship between socioeconomic deprivation and brain network topology.

View Article: PubMed Central - PubMed

Affiliation: Sackler Institute of Psychobiological Research, Institute of Health and Wellbeing, University of Glasgow, Gartnavel Royal Hospital Glasgow, UK.

ABSTRACT
Complex cognitive functions are widely recognized to be the result of a number of brain regions working together as large-scale networks. Recently, complex network analysis has been used to characterize various structural properties of the large-scale network organization of the brain. For example, the human brain has been found to have a modular architecture i.e., regions within the network form communities (modules) with more connections between regions within the community compared to regions outside it. The aim of this study was to examine the modular and overlapping modular architecture of the brain networks using complex network analysis. We also examined the association between neighborhood level deprivation and brain network structure-modularity and gray nodes. We compared network structure derived from anatomical MRI scans of 42 middle-aged neurologically healthy men from the least (LD) and the most deprived (MD) neighborhoods of Glasgow with their corresponding random networks. Cortical morphological covariance networks were constructed from the cortical thickness derived from the MRI scans of the brain. For a given modularity threshold, networks derived from the MD group showed similar number of modules compared to their corresponding random networks, while networks derived from the LD group had more modules compared to their corresponding random networks. The MD group also had fewer gray nodes-a measure of overlapping modular structure. These results suggest that apparent structural difference in brain networks may be driven by differences in cortical thicknesses between groups. This demonstrates a structural organization that is consistent with a system that is less robust and less efficient in information processing. These findings provide some evidence of the relationship between socioeconomic deprivation and brain network topology.

No MeSH data available.


Related in: MedlinePlus