Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


12695.pdf (4.3 MB)

ETD Abstract Container

Abstract Header

Method for Identifying Resting State Networks following Probabilistic Independent Component Analysis

Drake, David M
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1416231546

Abstract Details

2014, MS, University of Cincinnati, Engineering and Applied Science: Biomedical Engineering.
Independent Component Analysis (ICA) is a model-free functional connectivity technique which breaks down the blood oxygen level dependent signal into spatial maps (components) representing statistically independent fMRI source signals. Probabilistic ICA (PICA) improves upon the traditional ICA model by adding noise, and assumes for a non-square mixing matrix, enabling more accurate calculations of the spatial components. However, PICA still generates components in a random order requiring additional steps for identifying significant resting state networks (RSNs) across a group of subjects. The state-of-the-art for identifying these RSNs across a group of subjects following PICA either requires an extensive group-wise algorithm that derives individual component maps for each subject, or through correlating the components with a template to determine the goodness-of-fit. This thesis introduces an open-source template-matching algorithm for identifying resting state networks from components following any probabilistic Independent Component Analysis. Component selections are determined objectively through the use of correlations coefficients, and are excluded based on p-values calculated using Fisher’s R-to-Z transform. The final output of the algorithm displays both visually appealing and quantitative information enabling researchers to identify significant components without prior knowledge of their spatial distributions.
Jing-Huei Lee, Ph.D. (Committee Chair)
Wen-Jang Chu, Ph.D. (Committee Member)
Marepalli Rao, Ph.D. (Committee Member)
54 p.
Biomedical Research
Resting State Networks; Independent Component Analysis; Functional MRI; Correlation Coefficients; R programming; UNIX programming

Recommended Citations

Citations

  • Drake, D. M. (2014). Method for Identifying Resting State Networks following Probabilistic Independent Component Analysis [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1416231546

    APA Style (7th edition)

  • Drake, David. Method for Identifying Resting State Networks following Probabilistic Independent Component Analysis. 2014. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1416231546.

    MLA Style (8th edition)

  • Drake, David. "Method for Identifying Resting State Networks following Probabilistic Independent Component Analysis." Master's thesis, University of Cincinnati, 2014. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1416231546

    Chicago Manual of Style (17th edition)

ucin1416231546
683
© 2014, some rights reserved.Creative Commons License Method for Identifying Resting State Networks following Probabilistic Independent Component Analysis by David M Drake is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by University of Cincinnati and OhioLINK.