Loading...

Spectral clustering approach with sparsifying technique for functional connectivity detection in the resting brain

Ramezani, M ; Sharif University of Technology

873 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/ICIAS.2010.5716164
  3. Abstract:
  4. The aim of this study is to assess the functional connectivity from resting state functional magnetic resonance imaging (fMRI) data. Spectral clustering algorithm was applied to the realistic and real fMRI data acquired from a resting healthy subject to find functionally connected brain regions. In order to make computation of the spectral decompositions of the entire brain volume feasible, the similarity matrix has been sparsified with the t-nearestneighbor approach. Realistic data were created to investigate the performance of the proposed algorithm and comparing it to the recently proposed spectral clustering algorithm with the Nystrom approximation and also with some well-known algorithms such as the Cross Correlation Analysis (CCA) and the spatial Independent Component Analysis (sICA). To enhance the performance of the methods, a variety of data pre and post processing steps, including data normalization, outlier removal, dimensionality reduction by using wavelet coefficients, estimation of number of clusters and optimal number of independent components (ICs). Results demonstrate the applicability of the proposed algorithm for functional connectivity analysis
  5. Keywords:
  6. Brain regions ; Brain volume ; Cross-correlation analysis ; Data normalization ; Dimensionality reduction ; fMRI data ; Functional connectivity ; Functional magnetic resonance imaging ; Healthy subjects ; Independent components ; Nearest-neighbor approaches ; Number of clusters ; Optimal number ; Post processing ; Realistic data ; Resting state ; Similarity matrix ; Spatial independent component analysis ; Spectral clustering ; Spectral decomposition ; Wavelet coefficients ; Approximation algorithms ; Brain ; Cluster analysis ; Independent component analysis ; Magnetic resonance imaging ; Multivariant analysis ; Resonance ; Clustering algorithms
  7. Source: 2010 International Conference on Intelligent and Advanced Systems, ICIAS 2010, 15 June 2010 through 17 June 2010 ; 2010 ; 9781424466238 (ISBN)
  8. URL: http://ieeexplore.ieee.org/document/5716164