Weighted phase lag index and graph analysis: preliminary investigation of functional connectivity during resting state in children

Comput Math Methods Med. 2012:2012:186353. doi: 10.1155/2012/186353. Epub 2012 Sep 24.

Abstract

Resting state functional connectivity of MEG data was studied in 29 children (9-10 years old). The weighted phase lag index (WPLI) was employed for estimating connectivity and compared to coherence. To further evaluate the network structure, a graph analysis based on WPLI was used to determine clustering coefficient (C) and betweenness centrality (BC) as local coefficients as well as the characteristic path length (L) as a parameter for global interconnectedness. The network's modular structure was also calculated to estimate functional segregation. A seed region was identified in the central occipital area based on the power distribution at the sensor level in the alpha band. WPLI reveals a specific connectivity map different from power and coherence. BC and modularity show a strong level of connectedness in the occipital area between lateral and central sensors. C shows different isolated areas of occipital sensors. Globally, a network with the shortest L is detected in the alpha band, consistently with the local results. Our results are in agreement with findings in adults, indicating a similar functional network in children at this age in the alpha band. The integrated use of WPLI and graph analysis can help to gain a better description of resting state networks.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms
  • Brain Mapping / methods
  • Child
  • Cluster Analysis
  • Computational Biology / methods
  • Female
  • Humans
  • Magnetoencephalography / methods*
  • Male
  • Models, Neurological
  • Nerve Net / physiopathology
  • Neural Pathways
  • Regression Analysis
  • Reproducibility of Results
  • Rest
  • Signal Processing, Computer-Assisted