Effect of Spatial Alignment Transformations in PCA and ICA of Functional Neuroimages

Ana S. Lukic; Miles N. Wernick; Yongui Yang; Lars Kai Hansen; Konstantinos Arfanakis; Stephen C. Strother

doi:10.1109/TMI.2007.896928

Effect of Spatial Alignment Transformations in PCA and ICA of Functional Neuroimages

Ana S. Lukic, Miles N. Wernick, Yongui Yang, Lars Kai Hansen, Konstantinos Arfanakis, Stephen C. Strother

Research output: Contribution to journal › Journal article › Research › peer-review

428 Downloads (Pure)

Abstract

It has been previously observed that spatial independent component analysis (ICA), if applied to data pooled in a particular way, may lessen the need for spatial alignment of scans in a functional neuroimaging study. In this paper we seek to determine analytically the conditions under which this observation is true, not only for spatial ICA, but also for temporal ICA and for principal component analysis (PCA). In each case we find conditions that the spatial alignment operator must satisfy to ensure invariance of the results. We illustrate our findings using functional magnetic-resonance imaging (fMRI) data. Our analysis is applicable to both inter-subject and intra-subject spatial normalization.

Original language	English
Journal	I E E E Transactions on Medical Imaging
Volume	26
Issue number	8
Pages (from-to)	1058-1068
ISSN	0278-0062
DOIs	https://doi.org/10.1109/TMI.2007.896928
Publication status	Published - 2007

Keywords

Neuroimaging.
Independent Component Analysis
Image Registration
fMRI

Access to Document

10.1109/TMI.2007.896928

Imm5178Final published version, 764 KB

http://www2.imm.dtu.dk/pubdb/p.php?5178

OpenUrl availability

Full text

Cite this

@article{de6e2cffae3549ce9b0f6bee90a4d547,

title = "Effect of Spatial Alignment Transformations in PCA and ICA of Functional Neuroimages",

abstract = "It has been previously observed that spatial independent component analysis (ICA), if applied to data pooled in a particular way, may lessen the need for spatial alignment of scans in a functional neuroimaging study. In this paper we seek to determine analytically the conditions under which this observation is true, not only for spatial ICA, but also for temporal ICA and for principal component analysis (PCA). In each case we find conditions that the spatial alignment operator must satisfy to ensure invariance of the results. We illustrate our findings using functional magnetic-resonance imaging (fMRI) data. Our analysis is applicable to both inter-subject and intra-subject spatial normalization.",

keywords = "Neuroimaging., Independent Component Analysis, Image Registration, fMRI",

author = "Lukic, {Ana S.} and Wernick, {Miles N.} and Yongui Yang and Hansen, {Lars Kai} and Konstantinos Arfanakis and Strother, {Stephen C.}",

year = "2007",

doi = "10.1109/TMI.2007.896928",

language = "English",

volume = "26",

pages = "1058--1068",

journal = "I E E E Transactions on Medical Imaging",

issn = "0278-0062",

publisher = "Institute of Electrical and Electronics Engineers",

number = "8",

}

TY - JOUR

T1 - Effect of Spatial Alignment Transformations in PCA and ICA of Functional Neuroimages

AU - Lukic, Ana S.

AU - Wernick, Miles N.

AU - Yang, Yongui

AU - Hansen, Lars Kai

AU - Arfanakis, Konstantinos

AU - Strother, Stephen C.

PY - 2007

Y1 - 2007

N2 - It has been previously observed that spatial independent component analysis (ICA), if applied to data pooled in a particular way, may lessen the need for spatial alignment of scans in a functional neuroimaging study. In this paper we seek to determine analytically the conditions under which this observation is true, not only for spatial ICA, but also for temporal ICA and for principal component analysis (PCA). In each case we find conditions that the spatial alignment operator must satisfy to ensure invariance of the results. We illustrate our findings using functional magnetic-resonance imaging (fMRI) data. Our analysis is applicable to both inter-subject and intra-subject spatial normalization.

AB - It has been previously observed that spatial independent component analysis (ICA), if applied to data pooled in a particular way, may lessen the need for spatial alignment of scans in a functional neuroimaging study. In this paper we seek to determine analytically the conditions under which this observation is true, not only for spatial ICA, but also for temporal ICA and for principal component analysis (PCA). In each case we find conditions that the spatial alignment operator must satisfy to ensure invariance of the results. We illustrate our findings using functional magnetic-resonance imaging (fMRI) data. Our analysis is applicable to both inter-subject and intra-subject spatial normalization.

KW - Neuroimaging.

KW - Independent Component Analysis

KW - Image Registration

KW - fMRI

U2 - 10.1109/TMI.2007.896928

DO - 10.1109/TMI.2007.896928

M3 - Journal article

C2 - 17695126

SN - 0278-0062

VL - 26

SP - 1058

EP - 1068

JO - I E E E Transactions on Medical Imaging

JF - I E E E Transactions on Medical Imaging

IS - 8

ER -

Effect of Spatial Alignment Transformations in PCA and ICA of Functional Neuroimages

Abstract

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this