Image segmentation with the combination of the PCA and ICA-based modes of shape variation

Juha Koikkalainen; Jyrki Lötjönen

doi:10.1109/ISBI.2004.1398496

Image segmentation with the combination of the PCA and ICA-based modes of shape variation

Juha Koikkalainen, Jyrki Lötjönen

VTT Technical Research Centre of Finland

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

8 Citations (Scopus)

Abstract

How to constrain the deformations in deformable model based image segmentation is a well-studied issue. Many methods that use the modes of shape variation generated from a training set shapes have been introduced. Most of these methods rely on principle component analysis (PCA) to statistically model the variability in the training set. Independent component analysis (ICA) has been proposed for this purpose, too. In this paper, we combine the PCA- and ICA-based modes of shape variation using a consecutive approach: an a priori model is deformed first by the PCA modes, which represent the global shape variability in the training set, and then, by the ICA modes, which have a more local character. The method is validated using a set of three-dimensional (3-D) brain MR images. The results prove that by applying the ICA modes after the PCA modes the accuracy of image segmentation is statistically significantly (p < 0:05) improved.

Original language	English
Title of host publication	2004 2nd IEEE International Symposium on Biomedical Imaging
Subtitle of host publication	From Nano to Macro
Publisher	IEEE Institute of Electrical and Electronic Engineers
Pages	149-152
ISBN (Print)	978-0-7803-8389-0
DOIs	https://doi.org/10.1109/ISBI.2004.1398496
Publication status	Published - 2004
MoE publication type	A4 Article in a conference publication
Event	2nd IEEE International Symposium on Biomedical Imaging: From Nano to Macro - Arlington, United States Duration: 15 Apr 2004 → 18 Apr 2004

Conference

Conference	2nd IEEE International Symposium on Biomedical Imaging
Country/Territory	United States
City	Arlington
Period	15/04/04 → 18/04/04

Access to Document

10.1109/ISBI.2004.1398496

Cite this

@inproceedings{371367f347af40de80b477827b31c03b,

title = "Image segmentation with the combination of the PCA and ICA-based modes of shape variation",

abstract = "How to constrain the deformations in deformable model based image segmentation is a well-studied issue. Many methods that use the modes of shape variation generated from a training set shapes have been introduced. Most of these methods rely on principle component analysis (PCA) to statistically model the variability in the training set. Independent component analysis (ICA) has been proposed for this purpose, too. In this paper, we combine the PCA- and ICA-based modes of shape variation using a consecutive approach: an a priori model is deformed first by the PCA modes, which represent the global shape variability in the training set, and then, by the ICA modes, which have a more local character. The method is validated using a set of three-dimensional (3-D) brain MR images. The results prove that by applying the ICA modes after the PCA modes the accuracy of image segmentation is statistically significantly (p < 0:05) improved.",

author = "Juha Koikkalainen and Jyrki L{\"o}tj{\"o}nen",

note = "Project code: T4SU00131; 2nd IEEE International Symposium on Biomedical Imaging : From Nano to Macro ; Conference date: 15-04-2004 Through 18-04-2004",

year = "2004",

doi = "10.1109/ISBI.2004.1398496",

language = "English",

isbn = "978-0-7803-8389-0",

pages = "149--152",

booktitle = "2004 2nd IEEE International Symposium on Biomedical Imaging",

publisher = "IEEE Institute of Electrical and Electronic Engineers",

address = "United States",

}

Koikkalainen, J & Lötjönen, J 2004, Image segmentation with the combination of the PCA and ICA-based modes of shape variation. in 2004 2nd IEEE International Symposium on Biomedical Imaging: From Nano to Macro. IEEE Institute of Electrical and Electronic Engineers, pp. 149-152, 2nd IEEE International Symposium on Biomedical Imaging, Arlington, Virginia, United States, 15/04/04. https://doi.org/10.1109/ISBI.2004.1398496

Image segmentation with the combination of the PCA and ICA-based modes of shape variation. / Koikkalainen, Juha; Lötjönen, Jyrki.
2004 2nd IEEE International Symposium on Biomedical Imaging: From Nano to Macro. IEEE Institute of Electrical and Electronic Engineers, 2004. p. 149-152.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Image segmentation with the combination of the PCA and ICA-based modes of shape variation

AU - Koikkalainen, Juha

AU - Lötjönen, Jyrki

N1 - Project code: T4SU00131

PY - 2004

Y1 - 2004

N2 - How to constrain the deformations in deformable model based image segmentation is a well-studied issue. Many methods that use the modes of shape variation generated from a training set shapes have been introduced. Most of these methods rely on principle component analysis (PCA) to statistically model the variability in the training set. Independent component analysis (ICA) has been proposed for this purpose, too. In this paper, we combine the PCA- and ICA-based modes of shape variation using a consecutive approach: an a priori model is deformed first by the PCA modes, which represent the global shape variability in the training set, and then, by the ICA modes, which have a more local character. The method is validated using a set of three-dimensional (3-D) brain MR images. The results prove that by applying the ICA modes after the PCA modes the accuracy of image segmentation is statistically significantly (p < 0:05) improved.

AB - How to constrain the deformations in deformable model based image segmentation is a well-studied issue. Many methods that use the modes of shape variation generated from a training set shapes have been introduced. Most of these methods rely on principle component analysis (PCA) to statistically model the variability in the training set. Independent component analysis (ICA) has been proposed for this purpose, too. In this paper, we combine the PCA- and ICA-based modes of shape variation using a consecutive approach: an a priori model is deformed first by the PCA modes, which represent the global shape variability in the training set, and then, by the ICA modes, which have a more local character. The method is validated using a set of three-dimensional (3-D) brain MR images. The results prove that by applying the ICA modes after the PCA modes the accuracy of image segmentation is statistically significantly (p < 0:05) improved.

U2 - 10.1109/ISBI.2004.1398496

DO - 10.1109/ISBI.2004.1398496

M3 - Conference article in proceedings

SN - 978-0-7803-8389-0

SP - 149

EP - 152

BT - 2004 2nd IEEE International Symposium on Biomedical Imaging

PB - IEEE Institute of Electrical and Electronic Engineers

T2 - 2nd IEEE International Symposium on Biomedical Imaging

Y2 - 15 April 2004 through 18 April 2004

ER -

Image segmentation with the combination of the PCA and ICA-based modes of shape variation

Abstract

Conference

Access to Document

Fingerprint

Cite this