Brain response pattern identification of fMRI data using a particle swarm optimization-based approach

Xinpei Ma, Chun An Chou, Hiroki Sayama, Wanpracha Art Chaovalitwongse

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Many neuroscience studies have been devoted to understand brain neural responses correlating to cognition using functional magnetic resonance imaging (fMRI). In contrast to univariate analysis to identify response patterns, it is shown that multi-voxel pattern analysis (MVPA) of fMRI data becomes a relatively effective approach using machine learning techniques in the recent literature. MVPA can be considered as a multi-objective pattern classification problem with the aim to optimize response patterns, in which informative voxels interacting with each other are selected, achieving high classification accuracy associated with cognitive stimulus conditions. To solve the problem, we propose a feature interaction detection framework, integrating hierarchical heterogeneous particle swarm optimization and support vector machines, for voxel selection in MVPA. In the proposed approach, we first select the most informative voxels and then identify a response pattern based on the connectivity of the selected voxels. The effectiveness of the proposed approach was examined for the Haxby’s dataset of object-level representations. The computational results demonstrated higher classification accuracy by the extracted response patterns, compared to state-of-the-art feature selection algorithms, such as forward selection and backward selection.

Original languageEnglish
Pages (from-to)181-192
Number of pages12
JournalBrain Informatics
Volume3
Issue number3
DOIs
Publication statusPublished - 2016 Sep 1
Externally publishedYes

Fingerprint

Particle swarm optimization (PSO)
Brain
Magnetic Resonance Imaging
Pattern recognition
Support vector machines
Learning systems
Feature extraction
Neurosciences
Cognition
Support Vector Machine
Datasets
Machine Learning

Keywords

  • Brain functional connectivity
  • Brain response pattern
  • Feature selection
  • Interaction selection
  • Particle swarm optimization
  • Pattern classification

ASJC Scopus subject areas

  • Neurology
  • Computer Science Applications
  • Cognitive Neuroscience

Cite this

Brain response pattern identification of fMRI data using a particle swarm optimization-based approach. / Ma, Xinpei; Chou, Chun An; Sayama, Hiroki; Chaovalitwongse, Wanpracha Art.

In: Brain Informatics, Vol. 3, No. 3, 01.09.2016, p. 181-192.

Research output: Contribution to journalArticle

Ma, Xinpei ; Chou, Chun An ; Sayama, Hiroki ; Chaovalitwongse, Wanpracha Art. / Brain response pattern identification of fMRI data using a particle swarm optimization-based approach. In: Brain Informatics. 2016 ; Vol. 3, No. 3. pp. 181-192.
@article{a2c383c748fd49ac8780e1db370a42cd,
title = "Brain response pattern identification of fMRI data using a particle swarm optimization-based approach",
abstract = "Many neuroscience studies have been devoted to understand brain neural responses correlating to cognition using functional magnetic resonance imaging (fMRI). In contrast to univariate analysis to identify response patterns, it is shown that multi-voxel pattern analysis (MVPA) of fMRI data becomes a relatively effective approach using machine learning techniques in the recent literature. MVPA can be considered as a multi-objective pattern classification problem with the aim to optimize response patterns, in which informative voxels interacting with each other are selected, achieving high classification accuracy associated with cognitive stimulus conditions. To solve the problem, we propose a feature interaction detection framework, integrating hierarchical heterogeneous particle swarm optimization and support vector machines, for voxel selection in MVPA. In the proposed approach, we first select the most informative voxels and then identify a response pattern based on the connectivity of the selected voxels. The effectiveness of the proposed approach was examined for the Haxby’s dataset of object-level representations. The computational results demonstrated higher classification accuracy by the extracted response patterns, compared to state-of-the-art feature selection algorithms, such as forward selection and backward selection.",
keywords = "Brain functional connectivity, Brain response pattern, Feature selection, Interaction selection, Particle swarm optimization, Pattern classification",
author = "Xinpei Ma and Chou, {Chun An} and Hiroki Sayama and Chaovalitwongse, {Wanpracha Art}",
year = "2016",
month = "9",
day = "1",
doi = "10.1007/s40708-016-0049-z",
language = "English",
volume = "3",
pages = "181--192",
journal = "Brain Informatics",
issn = "2198-4018",
publisher = "Springer Berlin",
number = "3",

}

TY - JOUR

T1 - Brain response pattern identification of fMRI data using a particle swarm optimization-based approach

AU - Ma, Xinpei

AU - Chou, Chun An

AU - Sayama, Hiroki

AU - Chaovalitwongse, Wanpracha Art

PY - 2016/9/1

Y1 - 2016/9/1

N2 - Many neuroscience studies have been devoted to understand brain neural responses correlating to cognition using functional magnetic resonance imaging (fMRI). In contrast to univariate analysis to identify response patterns, it is shown that multi-voxel pattern analysis (MVPA) of fMRI data becomes a relatively effective approach using machine learning techniques in the recent literature. MVPA can be considered as a multi-objective pattern classification problem with the aim to optimize response patterns, in which informative voxels interacting with each other are selected, achieving high classification accuracy associated with cognitive stimulus conditions. To solve the problem, we propose a feature interaction detection framework, integrating hierarchical heterogeneous particle swarm optimization and support vector machines, for voxel selection in MVPA. In the proposed approach, we first select the most informative voxels and then identify a response pattern based on the connectivity of the selected voxels. The effectiveness of the proposed approach was examined for the Haxby’s dataset of object-level representations. The computational results demonstrated higher classification accuracy by the extracted response patterns, compared to state-of-the-art feature selection algorithms, such as forward selection and backward selection.

AB - Many neuroscience studies have been devoted to understand brain neural responses correlating to cognition using functional magnetic resonance imaging (fMRI). In contrast to univariate analysis to identify response patterns, it is shown that multi-voxel pattern analysis (MVPA) of fMRI data becomes a relatively effective approach using machine learning techniques in the recent literature. MVPA can be considered as a multi-objective pattern classification problem with the aim to optimize response patterns, in which informative voxels interacting with each other are selected, achieving high classification accuracy associated with cognitive stimulus conditions. To solve the problem, we propose a feature interaction detection framework, integrating hierarchical heterogeneous particle swarm optimization and support vector machines, for voxel selection in MVPA. In the proposed approach, we first select the most informative voxels and then identify a response pattern based on the connectivity of the selected voxels. The effectiveness of the proposed approach was examined for the Haxby’s dataset of object-level representations. The computational results demonstrated higher classification accuracy by the extracted response patterns, compared to state-of-the-art feature selection algorithms, such as forward selection and backward selection.

KW - Brain functional connectivity

KW - Brain response pattern

KW - Feature selection

KW - Interaction selection

KW - Particle swarm optimization

KW - Pattern classification

UR - http://www.scopus.com/inward/record.url?scp=85046027829&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046027829&partnerID=8YFLogxK

U2 - 10.1007/s40708-016-0049-z

DO - 10.1007/s40708-016-0049-z

M3 - Article

VL - 3

SP - 181

EP - 192

JO - Brain Informatics

JF - Brain Informatics

SN - 2198-4018

IS - 3

ER -