Between-group comparisons of structural and functional brain connectivity Open Access

Abstract

This thesis mainly focused on detecting the difference of brain connectivity between diseased vs. normal subjects in Philadelphia Neurodevelopmental Cohort (PNC) study. Multimodal neuroimaging including fMRI and diffusion MRI was used to investigate group differences in functional connectivity and structural connectivity with the brain. Probabilistic tractography was performed to estimate structural connectivity and partial correlations of fMRI time series data were used to estimate functional connectivity in subjects' brains. Edge-wise linear regression was then performed to detect the difference of structural and functional connectivity between two groups of subjects. Significant between-group differences were found in 105 region pairs for structural connectivity and in 264 regions pairs for functional connectivity. Two edge-wise linear regression were performed to detect the potential relationship between structural and functional connectivity for each group. More than 150 edges were detected having significant correlation between structural and functional connectivity. In the second part of the thesis, five graph theoretical metrics in graph theory were evaluated to show the network properties of structural and functional connectivity in each group. The linear regressions, which aim to detect the difference network properties between two groups of subjects, were performed. According to the linear models, characteristic path length had a significantly different between the diseased and normal subjects. Lastly, a recently developed clustering method, which is one of differentially expressed network methods, was performed to identify clusters of regions that showed significantly different structural and functional connectivity between two groups of subjects. We provided four clustered edges' networks for structural connectivity and functional connectivity to show the significant difference between two groups of subjects.

Table of Contents

1. Introduction. 1

2. Method. 4

2.1 Data Acquisition. 4

2.1.1 Subjects. 4

2.1.2 MRI data. 5

2.1.3 Data Pre-processing. 5

2.2 Probabilistic Tracktography. 7

2.3 Edge-wise Linear Regression. 9

2.3.1 Edge-wise Linear Regression for SC and FC. 10

2.3.2 Edge-wise relationship between SC and FC. 11

2.4 Global network metric-based method (GNM). 12

2.4.1 Graph theoretical metrics. 12

2.4.2 Linear Regression. 14

2.5 Differentially expressed network(DEN). 14

2.5.1 Parsimonious differential brain connectivity network detection Algorithm(Pard). 15

3. Results. 17

3.1 Edge-wise linear regression for SC and FC. 17

3.2 Relationship between SC and FC.. 18

3.3 Graph Theoretical metrics result. 19

3.4 Differentially expressed network result. 19

4. Discussion. 21

Reference. 23

5. Appendix. 25

A. Figure. 25

About this Master's Thesis

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Rollins School of Public Health
Department	Biostatistics and Bioinformatics
Subfield / Discipline	Biostatistics - MPH & MSPH
Degree	MSPH
Submission	Master's Thesis
Language	English
Research Field	Biology, Biostatistics Biology, Neuroscience Health Sciences, Public Health
Keyword	Structural Connectivity Between-group Comparisons Graphical Theory Functional Connectivity
Committee Chair / Thesis Advisor	Kundu, Suprateek , Emory University Guo, Ying, Emory University
Committee Members	Mayberg, Helen S, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Between-group comparisons of structural and functional brain connectivity ()	2018-08-28 10:30:46 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions