Artificial enlargement of a training set for statistical shape models

Jyrki Lötjönen; Kari Antila; Elina Lamminmäki; J. Koikkalainen; M. Lilja; T. Cootes

doi:10.1007/11494621_10

Artificial enlargement of a training set for statistical shape models

Jyrki Lötjönen, Kari Antila, Elina Lamminmäki, J. Koikkalainen, M. Lilja, T. Cootes

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

14 Citations (Scopus)

Abstract

Different methods were evaluated to enlarge artificially a training set which is used to build a statistical shape model. In this work, the shape model was built from MR data of 25 subjects and it consisted of ventricles, atria and epicardium. The method adding smooth non-rigid deformations to original training set examples produced the best results. The results indicated also that artificial deformation modes model better an unseen object than an equal number of standard PCA modes generated from original data.

Original language	English
Title of host publication	Functional Imaging and Modeling of the Heart
Subtitle of host publication	Third International Workshop, FIMH 2005
Publisher	Springer
Pages	92-101
ISBN (Electronic)	978-3-540-32081-4
ISBN (Print)	978-3-540-26161-2
DOIs	https://doi.org/10.1007/11494621_10
Publication status	Published - 2005
MoE publication type	A4 Article in a conference publication
Event	3rd International Workshop on Functional Imaging and Modeling of the Heart, FIMH 2005 - Barcelona, Spain Duration: 2 Jun 2005 → 4 Jun 2005 Conference number: 3

Publication series

Series	Lecture Notes in Computer Science
Volume	3504
ISSN	0302-9743

Conference

Conference	3rd International Workshop on Functional Imaging and Modeling of the Heart, FIMH 2005
Abbreviated title	FIMH 2005
Country/Territory	Spain
City	Barcelona
Period	2/06/05 → 4/06/05

Access to Document

10.1007/11494621_10

Cite this

@inproceedings{e8bd3385c9ec4aa7abe595dad983d455,

title = "Artificial enlargement of a training set for statistical shape models",

abstract = "Different methods were evaluated to enlarge artificially a training set which is used to build a statistical shape model. In this work, the shape model was built from MR data of 25 subjects and it consisted of ventricles, atria and epicardium. The method adding smooth non-rigid deformations to original training set examples produced the best results. The results indicated also that artificial deformation modes model better an unseen object than an equal number of standard PCA modes generated from original data.",

author = "Jyrki L{\"o}tj{\"o}nen and Kari Antila and Elina Lamminm{\"a}ki and J. Koikkalainen and M. Lilja and T. Cootes",

note = "PNA: Mattila, Elina M.; 3rd International Workshop on Functional Imaging and Modeling of the Heart, FIMH 2005, FIMH 2005 ; Conference date: 02-06-2005 Through 04-06-2005",

year = "2005",

doi = "10.1007/11494621_10",

language = "English",

isbn = "978-3-540-26161-2",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "92--101",

booktitle = "Functional Imaging and Modeling of the Heart",

address = "Germany",

}

Lötjönen, J, Antila, K, Lamminmäki, E, Koikkalainen, J, Lilja, M & Cootes, T 2005, Artificial enlargement of a training set for statistical shape models. in Functional Imaging and Modeling of the Heart: Third International Workshop, FIMH 2005. Springer, Lecture Notes in Computer Science, vol. 3504, pp. 92-101, 3rd International Workshop on Functional Imaging and Modeling of the Heart, FIMH 2005, Barcelona, Spain, 2/06/05. https://doi.org/10.1007/11494621_10

Artificial enlargement of a training set for statistical shape models. / Lötjönen, Jyrki; Antila, Kari; Lamminmäki, Elina et al.
Functional Imaging and Modeling of the Heart: Third International Workshop, FIMH 2005. Springer, 2005. p. 92-101 (Lecture Notes in Computer Science, Vol. 3504).

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

TY - GEN

T1 - Artificial enlargement of a training set for statistical shape models

AU - Lötjönen, Jyrki

AU - Antila, Kari

AU - Lamminmäki, Elina

AU - Koikkalainen, J.

AU - Lilja, M.

AU - Cootes, T.

N1 - Conference code: 3

PY - 2005

Y1 - 2005

N2 - Different methods were evaluated to enlarge artificially a training set which is used to build a statistical shape model. In this work, the shape model was built from MR data of 25 subjects and it consisted of ventricles, atria and epicardium. The method adding smooth non-rigid deformations to original training set examples produced the best results. The results indicated also that artificial deformation modes model better an unseen object than an equal number of standard PCA modes generated from original data.

AB - Different methods were evaluated to enlarge artificially a training set which is used to build a statistical shape model. In this work, the shape model was built from MR data of 25 subjects and it consisted of ventricles, atria and epicardium. The method adding smooth non-rigid deformations to original training set examples produced the best results. The results indicated also that artificial deformation modes model better an unseen object than an equal number of standard PCA modes generated from original data.

U2 - 10.1007/11494621_10

DO - 10.1007/11494621_10

M3 - Conference article in proceedings

SN - 978-3-540-26161-2

T3 - Lecture Notes in Computer Science

SP - 92

EP - 101

BT - Functional Imaging and Modeling of the Heart

PB - Springer

T2 - 3rd International Workshop on Functional Imaging and Modeling of the Heart, FIMH 2005

Y2 - 2 June 2005 through 4 June 2005

ER -

Artificial enlargement of a training set for statistical shape models

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this