Improving the spatial specificity of canonical correlation analysis in fMRI

Rajesh Ranjan Nandy, Dietmar Cordes

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

The contrast-to-noise ratio (CNR) is often very low in fMRI data, and standard univariate methods suffer from a loss of sensitivity in the context of noise. The increased power of a multivariate statistical analysis method known as canonical correlation analysis (CCA) in fMRI studies with low CNR was established previously. However, CCA in its conventional form has weak spatial specificity. In this work we propose a new assignment scheme to rectify this problem. It is shown that the new method has improved spatial specificity as well as sensitivity compared to conventional CCA for detecting activation patterns in fMRI.

Original languageEnglish
Pages (from-to)947-952
Number of pages6
JournalMagnetic Resonance in Medicine
Volume52
Issue number4
DOIs
StatePublished - 1 Oct 2004

Fingerprint

Noise
Magnetic Resonance Imaging
Activation Analysis
Multivariate Analysis

Keywords

  • CCA
  • Canonical correlation analysis
  • Contrast-to-noise ratio
  • Multivariate statistical analysis
  • Spatial specificity
  • fMRI

Cite this

@article{76191db4aa604abab3abdf82ccfed437,
title = "Improving the spatial specificity of canonical correlation analysis in fMRI",
abstract = "The contrast-to-noise ratio (CNR) is often very low in fMRI data, and standard univariate methods suffer from a loss of sensitivity in the context of noise. The increased power of a multivariate statistical analysis method known as canonical correlation analysis (CCA) in fMRI studies with low CNR was established previously. However, CCA in its conventional form has weak spatial specificity. In this work we propose a new assignment scheme to rectify this problem. It is shown that the new method has improved spatial specificity as well as sensitivity compared to conventional CCA for detecting activation patterns in fMRI.",
keywords = "CCA, Canonical correlation analysis, Contrast-to-noise ratio, Multivariate statistical analysis, Spatial specificity, fMRI",
author = "Nandy, {Rajesh Ranjan} and Dietmar Cordes",
year = "2004",
month = "10",
day = "1",
doi = "10.1002/mrm.20234",
language = "English",
volume = "52",
pages = "947--952",
journal = "Magnetic Resonance in Medicine",
issn = "0740-3194",
publisher = "John Wiley and Sons Inc.",
number = "4",

}

Improving the spatial specificity of canonical correlation analysis in fMRI. / Nandy, Rajesh Ranjan; Cordes, Dietmar.

In: Magnetic Resonance in Medicine, Vol. 52, No. 4, 01.10.2004, p. 947-952.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Improving the spatial specificity of canonical correlation analysis in fMRI

AU - Nandy, Rajesh Ranjan

AU - Cordes, Dietmar

PY - 2004/10/1

Y1 - 2004/10/1

N2 - The contrast-to-noise ratio (CNR) is often very low in fMRI data, and standard univariate methods suffer from a loss of sensitivity in the context of noise. The increased power of a multivariate statistical analysis method known as canonical correlation analysis (CCA) in fMRI studies with low CNR was established previously. However, CCA in its conventional form has weak spatial specificity. In this work we propose a new assignment scheme to rectify this problem. It is shown that the new method has improved spatial specificity as well as sensitivity compared to conventional CCA for detecting activation patterns in fMRI.

AB - The contrast-to-noise ratio (CNR) is often very low in fMRI data, and standard univariate methods suffer from a loss of sensitivity in the context of noise. The increased power of a multivariate statistical analysis method known as canonical correlation analysis (CCA) in fMRI studies with low CNR was established previously. However, CCA in its conventional form has weak spatial specificity. In this work we propose a new assignment scheme to rectify this problem. It is shown that the new method has improved spatial specificity as well as sensitivity compared to conventional CCA for detecting activation patterns in fMRI.

KW - CCA

KW - Canonical correlation analysis

KW - Contrast-to-noise ratio

KW - Multivariate statistical analysis

KW - Spatial specificity

KW - fMRI

UR - http://www.scopus.com/inward/record.url?scp=4744359266&partnerID=8YFLogxK

U2 - 10.1002/mrm.20234

DO - 10.1002/mrm.20234

M3 - Article

C2 - 15389937

AN - SCOPUS:4744359266

VL - 52

SP - 947

EP - 952

JO - Magnetic Resonance in Medicine

JF - Magnetic Resonance in Medicine

SN - 0740-3194

IS - 4

ER -