Apply Group ICA Method into fMRI Analysis Using GIFT Package and Regression Models Open Access

Abstract

This paper focused on the application of group independent component analysis (ICA) method into fMRI analysis and evaluating covariate effects on independent components in group ICA. We first applied GIFT package developed by medical image analysis laboratory (MIALab) for group ICA analysis on our dataset. To evaluate the covariate effects, we considered the Mancovan package in GIFT. Our major issue with Mancovan is that it works on group IC maps instead of subject-specific IC maps for evaluating the covariate effects. To address this limitation, we propose a multivariate regression method based on subject-specific maps. We evaluated the performance of our regression approach using simulation studies that generated multi-subject fMRI data with simulated covariates. The accuracy of the method is evaluated by the Mean square error (MSE) based on the difference between the true and estimated covariate parameters. Our results show that the proposed regression method may provide a good way to assess covariate effects on spatial distributions of ICs extracted from fMRI data.

Table of Contents

Chapter I

Introduction………………………………………………………………………………………………………1

Overview and background………………………………………………………………………………….1

Organization of human brain………………………………………………………………………………1

fMRI……………………………………………………………………………………………………………………2

fMRI data……………………………………………………………………………………………………………3

Chapter II

Methodology and Experiment…………………………………………………………………………..5

Review of ICA methods……………………………………………………………………………………….5

Introduction to group ICA methods…………………………………………………………………….6

Description of data……………………………………………………………………………………………..9

Experiment procedure……………………………………………………………………………………….10

Simulation study………………………………………………………………………………………………..13

Chapter III

Results and Conclusions……………………………………………………………………………………16

Chapter IV

Discussion………………………………………………………………………………………………………….18

Reference……………………………………………………………………………………………………………………21

Appendix A………………………………………………………………………………………………………………….22

Appendix B..................................................................................................23

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
Keyword	Regression fMRI ICA
Committee Chair / Thesis Advisor	Guo, Ying, Emory University
Committee Members	Kang, Jian , Emory University
Partnering Agencies	Does not apply (no collaborating organization)

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Apply Group ICA Method into fMRI Analysis Using GIFT Package and Regression Models ()	2018-08-28 12:34:36 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions