Loading...

EEG Brain Functional Network Analysis in Cortex Level

Pedrood, Bahman | 2013

606 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 44049 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Jalili, Mahdi
  7. Abstract:
  8. Complex networks science have received tremendous attention in recent years and the brain is one of the systems to which graph theoretical tools have been applied. Alzheimer’s disease (AD) is a neurodegenerative disease affecting many of elderly population. AD changes the anatomy of the brain, which subsequently results in changes in its functions. These changes have been frequently reported in signals recorded from the brain (such as MEG, fMRI and EEG). Among these neuroimaging techniques EEG is one of the most aproprate methods for extracting functional connectivites according to high temporal resolution. In this thesis, we aimed at analyzing the properties of EEG-based functional networks in AD. We considered EEG data of 26 AD patients and 26 healthy subjects in eyes-opened resting state. To have more accurate and informative representation of data, time series in the source level was obtained and was used for extracting the topology of functional networks. A number of neurobiologically meaningful graph theory metrics (including the global efficiency, clustering coefficient, small-worldness, assortativity and modularity) were computed for the networks. While no singnificant difference could be found in computed metrics of sensor level networks between groups of patients and healthy subjects, using source level networks surprised us by showing bunch of significant differences. AD patients mainly showed close-to-normal global connectivity – as measured by the global efficiency – and altered local connectivity – as measured by clustering coefficient and modularity. Indeed, they showed loss of small-worldness as compared to normal subjects. Furthermore, AD patients showed significantly different assortativity than the healthy subjects. On the other hand, hub nodes shown strong resiliency against AD by means of no significant changes in their number and locations
  9. Keywords:
  10. Brain Source Localization ; Alzheimer ; Network Analysis ; Complex Network ; Electroencephalography ; Brain Functional Network

 Digital Object List

 Bookmark

No TOC