Limits...
A study of brain networks associated with swallowing using graph-theoretical approaches.

Luan B, Sörös P, Sejdić E - PLoS ONE (2013)

Bottom Line: The results have shown that significant pairwise functional connections were, mostly, either local and intra-hemispheric or symmetrically inter-hemispheric.Furthermore, we showed that all human brain functional network, although varying in some degree, had typical small-world properties as compared to regular networks and random networks.Swallowing-related brain regions also had higher values for some of the network measures in comparison to when these measures were calculated for the whole brain.

View Article: PubMed Central - PubMed

Affiliation: Department of Electrical and Computer Engineering, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA.

ABSTRACT
Functional connectivity between brain regions during swallowing tasks is still not well understood. Understanding these complex interactions is of great interest from both a scientific and a clinical perspective. In this study, functional magnetic resonance imaging (fMRI) was utilized to study brain functional networks during voluntary saliva swallowing in twenty-two adult healthy subjects (all females, [Formula: see text] years of age). To construct these functional connections, we computed mean partial correlation matrices over ninety brain regions for each participant. Two regions were determined to be functionally connected if their correlation was above a certain threshold. These correlation matrices were then analyzed using graph-theoretical approaches. In particular, we considered several network measures for the whole brain and for swallowing-related brain regions. The results have shown that significant pairwise functional connections were, mostly, either local and intra-hemispheric or symmetrically inter-hemispheric. Furthermore, we showed that all human brain functional network, although varying in some degree, had typical small-world properties as compared to regular networks and random networks. These properties allow information transfer within the network at a relatively high efficiency. Swallowing-related brain regions also had higher values for some of the network measures in comparison to when these measures were calculated for the whole brain. The current results warrant further investigation of graph-theoretical approaches as a potential tool for understanding the neural basis of dysphagia.

Show MeSH

Related in: MedlinePlus

Comparison of networks measures for the swallowing ROIs and the whole brain: (a) global efficiency  (b) characteristic path length  (c) node degree  (d) clustering coefficient  (e) mean local efficiency  (f) hierarchy .
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3756959&req=5

pone-0073577-g003: Comparison of networks measures for the swallowing ROIs and the whole brain: (a) global efficiency (b) characteristic path length (c) node degree (d) clustering coefficient (e) mean local efficiency (f) hierarchy .

Mentions: Figure 3 demonstrated significant differences between the whole-brain matrices and swallowing ROIs for some of the network properties. No obvious difference in node degree was discovered between the two groups (). However, global efficiency was higher when considering swallowing ROIs and sparsity threshold values lower than 0.35, but it did not reach statistical significance for all values (). The path length of the binary and weighted network were significantly shorter in the whole-brain metric compared to swallowing related regions () when the threshold value was within the range of 0.60 to 0.85. The local efficiency values were significantly higher when considering swallowing ROIs and threshold values within the range of 0 to 0.03 (). Interestingly, we found that clustering coefficient value has slightly increased when we applied thresholds between 0.5 to 0.63. The rank-sum test showed that significant differences () had been found when comparing the whole brain and swallowing ROIs. The observed differences in the clustering coefficient were even greater in this interval in comparison to low threshold values. This has never been found in other network measurement parameters. As shown in Figure 3 (f), the hierarchy values for swallowing ROIs and the whole brain were not statistically different ().


A study of brain networks associated with swallowing using graph-theoretical approaches.

Luan B, Sörös P, Sejdić E - PLoS ONE (2013)

Comparison of networks measures for the swallowing ROIs and the whole brain: (a) global efficiency  (b) characteristic path length  (c) node degree  (d) clustering coefficient  (e) mean local efficiency  (f) hierarchy .
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3756959&req=5

pone-0073577-g003: Comparison of networks measures for the swallowing ROIs and the whole brain: (a) global efficiency (b) characteristic path length (c) node degree (d) clustering coefficient (e) mean local efficiency (f) hierarchy .
Mentions: Figure 3 demonstrated significant differences between the whole-brain matrices and swallowing ROIs for some of the network properties. No obvious difference in node degree was discovered between the two groups (). However, global efficiency was higher when considering swallowing ROIs and sparsity threshold values lower than 0.35, but it did not reach statistical significance for all values (). The path length of the binary and weighted network were significantly shorter in the whole-brain metric compared to swallowing related regions () when the threshold value was within the range of 0.60 to 0.85. The local efficiency values were significantly higher when considering swallowing ROIs and threshold values within the range of 0 to 0.03 (). Interestingly, we found that clustering coefficient value has slightly increased when we applied thresholds between 0.5 to 0.63. The rank-sum test showed that significant differences () had been found when comparing the whole brain and swallowing ROIs. The observed differences in the clustering coefficient were even greater in this interval in comparison to low threshold values. This has never been found in other network measurement parameters. As shown in Figure 3 (f), the hierarchy values for swallowing ROIs and the whole brain were not statistically different ().

Bottom Line: The results have shown that significant pairwise functional connections were, mostly, either local and intra-hemispheric or symmetrically inter-hemispheric.Furthermore, we showed that all human brain functional network, although varying in some degree, had typical small-world properties as compared to regular networks and random networks.Swallowing-related brain regions also had higher values for some of the network measures in comparison to when these measures were calculated for the whole brain.

View Article: PubMed Central - PubMed

Affiliation: Department of Electrical and Computer Engineering, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA.

ABSTRACT
Functional connectivity between brain regions during swallowing tasks is still not well understood. Understanding these complex interactions is of great interest from both a scientific and a clinical perspective. In this study, functional magnetic resonance imaging (fMRI) was utilized to study brain functional networks during voluntary saliva swallowing in twenty-two adult healthy subjects (all females, [Formula: see text] years of age). To construct these functional connections, we computed mean partial correlation matrices over ninety brain regions for each participant. Two regions were determined to be functionally connected if their correlation was above a certain threshold. These correlation matrices were then analyzed using graph-theoretical approaches. In particular, we considered several network measures for the whole brain and for swallowing-related brain regions. The results have shown that significant pairwise functional connections were, mostly, either local and intra-hemispheric or symmetrically inter-hemispheric. Furthermore, we showed that all human brain functional network, although varying in some degree, had typical small-world properties as compared to regular networks and random networks. These properties allow information transfer within the network at a relatively high efficiency. Swallowing-related brain regions also had higher values for some of the network measures in comparison to when these measures were calculated for the whole brain. The current results warrant further investigation of graph-theoretical approaches as a potential tool for understanding the neural basis of dysphagia.

Show MeSH
Related in: MedlinePlus