Performance evaluation of nonnegative matrix factorization algorithms to estimate task-related neuronal activities from fMRI data

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Nonnegative matrix factorization (NMF) is a blind source separation (BSS) algorithm which is based on the distinct constraint of nonnegativity of the estimated parameters as well as on the measured data. In this study, according to the potential feasibility of NMF for fMRI data, the four most popular NMF algorithms, corresponding to the following two types of (1) least-squares based update [i.e., alternating least-squares NMF (ALSNMF) and projected gradient descent NMF] and (2) multiplicative update (i.e., NMF based on Euclidean distance and NMF based on divergence cost function), were investigated by using them to estimate task-related neuronal activities. These algorithms were applied firstly to individual data from a single subject and, subsequently, to group data sets from multiple subjects. On the single-subject level, although all four algorithms detected task-related activation from simulated data, the performance of multiplicative update NMFs was significantly deteriorated when evaluated using visuomotor task fMRI data, for which they failed in estimating any task-related neuronal activities. In group-level analysis on both simulated data and real fMRI data, ALSNMF outperformed the other three algorithms. The presented findings may suggest that ALSNMF appears to be the most promising option among the tested NMF algorithms to extract task-related neuronal activities from fMRI data.

Original languageEnglish
Pages (from-to)466-476
Number of pages11
JournalMagnetic Resonance Imaging
Volume31
Issue number3
DOIs
Publication statusPublished - 2013 Apr 1

Fingerprint

Factorization
Magnetic Resonance Imaging
Least-Squares Analysis
Blind source separation
Cost functions
Costs and Cost Analysis
Chemical activation

Keywords

  • Blind source separation (BSS)
  • Functional magnetic resonance imaging (fMRI)
  • Nonnegative matrix factorization (NMF)
  • Visuomotor task

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging
  • Biomedical Engineering

Cite this

@article{322790d07cc943799a5c1e58ce9d2f3e,
title = "Performance evaluation of nonnegative matrix factorization algorithms to estimate task-related neuronal activities from fMRI data",
abstract = "Nonnegative matrix factorization (NMF) is a blind source separation (BSS) algorithm which is based on the distinct constraint of nonnegativity of the estimated parameters as well as on the measured data. In this study, according to the potential feasibility of NMF for fMRI data, the four most popular NMF algorithms, corresponding to the following two types of (1) least-squares based update [i.e., alternating least-squares NMF (ALSNMF) and projected gradient descent NMF] and (2) multiplicative update (i.e., NMF based on Euclidean distance and NMF based on divergence cost function), were investigated by using them to estimate task-related neuronal activities. These algorithms were applied firstly to individual data from a single subject and, subsequently, to group data sets from multiple subjects. On the single-subject level, although all four algorithms detected task-related activation from simulated data, the performance of multiplicative update NMFs was significantly deteriorated when evaluated using visuomotor task fMRI data, for which they failed in estimating any task-related neuronal activities. In group-level analysis on both simulated data and real fMRI data, ALSNMF outperformed the other three algorithms. The presented findings may suggest that ALSNMF appears to be the most promising option among the tested NMF algorithms to extract task-related neuronal activities from fMRI data.",
keywords = "Blind source separation (BSS), Functional magnetic resonance imaging (fMRI), Nonnegative matrix factorization (NMF), Visuomotor task",
author = "Xiaoyu Ding and Jong-Hwan Lee and Lee, {Seong Whan}",
year = "2013",
month = "4",
day = "1",
doi = "10.1016/j.mri.2012.10.003",
language = "English",
volume = "31",
pages = "466--476",
journal = "Magnetic Resonance Imaging",
issn = "0730-725X",
publisher = "Elsevier Inc.",
number = "3",

}

TY - JOUR

T1 - Performance evaluation of nonnegative matrix factorization algorithms to estimate task-related neuronal activities from fMRI data

AU - Ding, Xiaoyu

AU - Lee, Jong-Hwan

AU - Lee, Seong Whan

PY - 2013/4/1

Y1 - 2013/4/1

N2 - Nonnegative matrix factorization (NMF) is a blind source separation (BSS) algorithm which is based on the distinct constraint of nonnegativity of the estimated parameters as well as on the measured data. In this study, according to the potential feasibility of NMF for fMRI data, the four most popular NMF algorithms, corresponding to the following two types of (1) least-squares based update [i.e., alternating least-squares NMF (ALSNMF) and projected gradient descent NMF] and (2) multiplicative update (i.e., NMF based on Euclidean distance and NMF based on divergence cost function), were investigated by using them to estimate task-related neuronal activities. These algorithms were applied firstly to individual data from a single subject and, subsequently, to group data sets from multiple subjects. On the single-subject level, although all four algorithms detected task-related activation from simulated data, the performance of multiplicative update NMFs was significantly deteriorated when evaluated using visuomotor task fMRI data, for which they failed in estimating any task-related neuronal activities. In group-level analysis on both simulated data and real fMRI data, ALSNMF outperformed the other three algorithms. The presented findings may suggest that ALSNMF appears to be the most promising option among the tested NMF algorithms to extract task-related neuronal activities from fMRI data.

AB - Nonnegative matrix factorization (NMF) is a blind source separation (BSS) algorithm which is based on the distinct constraint of nonnegativity of the estimated parameters as well as on the measured data. In this study, according to the potential feasibility of NMF for fMRI data, the four most popular NMF algorithms, corresponding to the following two types of (1) least-squares based update [i.e., alternating least-squares NMF (ALSNMF) and projected gradient descent NMF] and (2) multiplicative update (i.e., NMF based on Euclidean distance and NMF based on divergence cost function), were investigated by using them to estimate task-related neuronal activities. These algorithms were applied firstly to individual data from a single subject and, subsequently, to group data sets from multiple subjects. On the single-subject level, although all four algorithms detected task-related activation from simulated data, the performance of multiplicative update NMFs was significantly deteriorated when evaluated using visuomotor task fMRI data, for which they failed in estimating any task-related neuronal activities. In group-level analysis on both simulated data and real fMRI data, ALSNMF outperformed the other three algorithms. The presented findings may suggest that ALSNMF appears to be the most promising option among the tested NMF algorithms to extract task-related neuronal activities from fMRI data.

KW - Blind source separation (BSS)

KW - Functional magnetic resonance imaging (fMRI)

KW - Nonnegative matrix factorization (NMF)

KW - Visuomotor task

UR - http://www.scopus.com/inward/record.url?scp=84875262877&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875262877&partnerID=8YFLogxK

U2 - 10.1016/j.mri.2012.10.003

DO - 10.1016/j.mri.2012.10.003

M3 - Article

C2 - 23200679

AN - SCOPUS:84875262877

VL - 31

SP - 466

EP - 476

JO - Magnetic Resonance Imaging

JF - Magnetic Resonance Imaging

SN - 0730-725X

IS - 3

ER -