Estimation of the intrinsic dimensionality of fMRI data

Dietmar Cordes; Rajesh R. Nandy

doi:10.1016/j.neuroimage.2005.07.054

Estimation of the intrinsic dimensionality of fMRI data

Dietmar Cordes, Rajesh R. Nandy

Research output: Contribution to journal › Article › peer-review

49 Scopus citations

Abstract

A new method based on an autoregressive noise model of order 1 is introduced to the problem of detecting the number of components in fMRI data. Unlike current information-theoretic criteria like AIC, MDL, and related PPCA, which do not incorporate autocorrelations in the noise, the new method leads to more consistent estimates of the model order, as illustrated in simulated and real fMRI resting-state data.

Original language	English
Pages (from-to)	145-154
Number of pages	10
Journal	NeuroImage
Volume	29
Issue number	1
DOIs	https://doi.org/10.1016/j.neuroimage.2005.07.054
State	Published - 1 Jan 2006

Keywords

Data reduction
Dimensionality estimation
PCA, Principal Component Analysis
fMRI, functional Magnetic Resonance Imaging

Access to Document

10.1016/j.neuroimage.2005.07.054

Cite this

@article{2ee3f78bfbdd43778c7f7be08d2b69fb,

title = "Estimation of the intrinsic dimensionality of fMRI data",

abstract = "A new method based on an autoregressive noise model of order 1 is introduced to the problem of detecting the number of components in fMRI data. Unlike current information-theoretic criteria like AIC, MDL, and related PPCA, which do not incorporate autocorrelations in the noise, the new method leads to more consistent estimates of the model order, as illustrated in simulated and real fMRI resting-state data.",

keywords = "Data reduction, Dimensionality estimation, PCA, Principal Component Analysis, fMRI, functional Magnetic Resonance Imaging",

author = "Dietmar Cordes and Nandy, {Rajesh R.}",

note = "Funding Information: The study was partially supported by grant P5033812 from the National Institute of Child Health and Human Development (NICHD) and a grant from the Royalty Research Fund of the University of Washington. ",

year = "2006",

month = jan,

day = "1",

doi = "10.1016/j.neuroimage.2005.07.054",

language = "English",

volume = "29",

pages = "145--154",

journal = "NeuroImage",

issn = "1053-8119",

publisher = "Academic Press Inc.",

number = "1",

}

TY - JOUR

T1 - Estimation of the intrinsic dimensionality of fMRI data

AU - Cordes, Dietmar

AU - Nandy, Rajesh R.

N1 - Funding Information: The study was partially supported by grant P5033812 from the National Institute of Child Health and Human Development (NICHD) and a grant from the Royalty Research Fund of the University of Washington.

PY - 2006/1/1

Y1 - 2006/1/1

N2 - A new method based on an autoregressive noise model of order 1 is introduced to the problem of detecting the number of components in fMRI data. Unlike current information-theoretic criteria like AIC, MDL, and related PPCA, which do not incorporate autocorrelations in the noise, the new method leads to more consistent estimates of the model order, as illustrated in simulated and real fMRI resting-state data.

AB - A new method based on an autoregressive noise model of order 1 is introduced to the problem of detecting the number of components in fMRI data. Unlike current information-theoretic criteria like AIC, MDL, and related PPCA, which do not incorporate autocorrelations in the noise, the new method leads to more consistent estimates of the model order, as illustrated in simulated and real fMRI resting-state data.

KW - Data reduction

KW - Dimensionality estimation

KW - PCA, Principal Component Analysis

KW - fMRI, functional Magnetic Resonance Imaging

UR - http://www.scopus.com/inward/record.url?scp=29244443235&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2005.07.054

DO - 10.1016/j.neuroimage.2005.07.054

M3 - Article

C2 - 16202626

AN - SCOPUS:29244443235

SN - 1053-8119

VL - 29

SP - 145

EP - 154

JO - NeuroImage

JF - NeuroImage

IS - 1

ER -

Estimation of the intrinsic dimensionality of fMRI data

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this