Cargando…
Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering
Human brain functional system has been viewed as a complex network. To accurately characterize this brain network, it is important to estimate the functional connectivity between separate brain regions (i.e., association matrix). One common approach to evaluating the connectivity is the pairwise Pea...
Autores principales: | , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2015
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4607787/ https://www.ncbi.nlm.nih.gov/pubmed/26528123 http://dx.doi.org/10.3389/fnins.2015.00383 |
_version_ | 1782395554480783360 |
---|---|
author | Li, Xuan Wang, Haixian |
author_facet | Li, Xuan Wang, Haixian |
author_sort | Li, Xuan |
collection | PubMed |
description | Human brain functional system has been viewed as a complex network. To accurately characterize this brain network, it is important to estimate the functional connectivity between separate brain regions (i.e., association matrix). One common approach to evaluating the connectivity is the pairwise Pearson correlation. However, this bivariate method completely ignores the influence of other regions when computing the pairwise association. Another intractable issue existed in many approaches to further analyzing the network structure is the requirement of applying a threshold to the association matrix. To address these issues, we develop a novel scheme to investigate the brain functional networks. Specifically, we first establish a global functional connection network by using the Adaptive Sparse Representation (ASR), adaptively integrating the sparsity of ℓ(1)-norm and the grouping effect of ℓ(2)-norm for linear representation and then identify connectivity patterns with Affinity Propagation (AP) clustering algorithm. Results on both simulated and real data indicate that the proposed scheme is superior to the Pearson correlation in connectivity quality and clustering quality. Our findings suggest that the proposed scheme is an accurate and useful technique to delineate functional network structure for functionally parsimonious and correlated fMRI data with a large number of brain regions. |
format | Online Article Text |
id | pubmed-4607787 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2015 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-46077872015-11-02 Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering Li, Xuan Wang, Haixian Front Neurosci Neuroscience Human brain functional system has been viewed as a complex network. To accurately characterize this brain network, it is important to estimate the functional connectivity between separate brain regions (i.e., association matrix). One common approach to evaluating the connectivity is the pairwise Pearson correlation. However, this bivariate method completely ignores the influence of other regions when computing the pairwise association. Another intractable issue existed in many approaches to further analyzing the network structure is the requirement of applying a threshold to the association matrix. To address these issues, we develop a novel scheme to investigate the brain functional networks. Specifically, we first establish a global functional connection network by using the Adaptive Sparse Representation (ASR), adaptively integrating the sparsity of ℓ(1)-norm and the grouping effect of ℓ(2)-norm for linear representation and then identify connectivity patterns with Affinity Propagation (AP) clustering algorithm. Results on both simulated and real data indicate that the proposed scheme is superior to the Pearson correlation in connectivity quality and clustering quality. Our findings suggest that the proposed scheme is an accurate and useful technique to delineate functional network structure for functionally parsimonious and correlated fMRI data with a large number of brain regions. Frontiers Media S.A. 2015-10-16 /pmc/articles/PMC4607787/ /pubmed/26528123 http://dx.doi.org/10.3389/fnins.2015.00383 Text en Copyright © 2015 Li and Wang. http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
spellingShingle | Neuroscience Li, Xuan Wang, Haixian Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering |
title | Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering |
title_full | Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering |
title_fullStr | Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering |
title_full_unstemmed | Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering |
title_short | Identification of functional networks in resting state fMRI data using adaptive sparse representation and affinity propagation clustering |
title_sort | identification of functional networks in resting state fmri data using adaptive sparse representation and affinity propagation clustering |
topic | Neuroscience |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4607787/ https://www.ncbi.nlm.nih.gov/pubmed/26528123 http://dx.doi.org/10.3389/fnins.2015.00383 |
work_keys_str_mv | AT lixuan identificationoffunctionalnetworksinrestingstatefmridatausingadaptivesparserepresentationandaffinitypropagationclustering AT wanghaixian identificationoffunctionalnetworksinrestingstatefmridatausingadaptivesparserepresentationandaffinitypropagationclustering |