Back to search

University of British Columbia

Connectivity-based parcellation of putamen region using resting state fMRI

Abstract

dc:description

Functional magnetic resonance imaging (fMRI) has shown great potential in studying the underlying neural systems. Functional connectivity measured by fMRI provides an efficient approach to study the interactions and relationships between different brain regions. However, functional connectivity studies require accurate definition of brain regions, which is often difficult and may not be achieved through anatomical landmarks. In this thesis, we present a novel framework for parcellation of a brain region into functional subunits based on their connectivity patterns with other reference brain regions. The proposed method takes the prior neurological information into consideration and aims at finding spatially continuous and functionally consistent sub-regions in a given brain region. The proposed framework relies on a sparse spatially regularized fused lasso regression model for feature extraction. The usual lasso model is a linear regression model commonly applied in high dimensional data such as fMRI signals. Compared with lasso, the proposed model further considers the spatial order of each voxel and thus encourages spatially and functionally adjacent voxels to share similar regression coefficients despite of the possible spatial noise. In order to achieve the accurate parcellation results, we propose a process by iteratively merging voxels (groups) and tuning the parameters adaptively. In addition, a Graph-Cut optimization algorithm is adopted for assigning the overlapped voxels into separate sub-regions. With spatial information incorporated, spatially continuous and functionally consistent subunits can be obtained which are desired for subsequent brain connectivity analysis. The simulation results demonstrate that the proposed method could reliably yield spatially continuous and functionally consistent subunits. When applied to real resting state fMRI datasets, two consistent functional subunits could be obtained in the putamen region for all normal subjects. Comparisons between the results of the Parkinson’s disease group and the normal group suggest that the obtained results are in accordance with our medical assumption. The extracted functional subunits themselves are of great interest in studying the influence of aging and a certain disease, and they may provide us deeper insights and serve as a biomarker in our future Parkinson’s disease study.

Degree

thesis:*
Name thesis:degree_name
Master of Applied Science - MASc
Level thesis:degree_level
master's
Discipline thesis:degree_discipline
Electrical and Computer Engineering
Grantor dc:publisher
University of British Columbia
Year dc:date
2015

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Zhang, Yiming

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivs 2.5 Canada
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/53970

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Zhang, Yiming. Connectivity-based parcellation of putamen region using resting state fMRI. master's thesis, University of British Columbia, 2015. http://hdl.handle.net/2429/53970