Asymmetric image-template registration

Mert R. Sabuncu; B. T. Thomas Yeo; Koen Van Leemput; Tom Vercauteren; Polina Golland

doi:10.1007/978-3-642-04268-3_70

Asymmetric image-template registration

Mert R. Sabuncu, B. T. Thomas Yeo, Koen Van Leemput, Tom Vercauteren, Polina Golland

Massachusetts Institute of Technology

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

Abstract

A natural requirement in pairwise image registration is that the resulting deformation is independent of the order of the images. This constraint is typically achieved via a symmetric cost function and has been shown to reduce the effects of local optima. Consequently, symmetric registration has been successfully applied to pairwise image registration as well as the spatial alignment of individual images with a template. However, recent work has shown that the relationship between an image and a template is fundamentally asymmetric. In this paper, we develop a method that reconciles the practical advantages of symmetric registration with the asymmetric nature of image-template registration by adding a simple correction factor to the symmetric cost function. We instantiate our model within a log-domain diffeomorphic registration framework. Our experiments show exploiting the asymmetry in image-template registration improves alignment in the image coordinates. © 2009 Springer-Verlag.

Original language	English
Title of host publication	Medical Image Computing and Computer-Assisted Intervention
Volume	5761
Publisher	Springer-verlag Berlin
Publication date	2009
Pages	565-573
Chapter	9
ISBN (Print)	978-3-642-04267-6
DOIs	https://doi.org/10.1007/978-3-642-04268-3_70
Publication status	Published - 2009
Externally published	Yes
Event	12th International Conference on Medical Image Computing and Computer-Assisted Intervention - Imperial College, London, United Kingdom Duration: 20 Sept 2009 → 24 Sept 2009 Conference number: 12

Conference

Conference	12th International Conference on Medical Image Computing and Computer-Assisted Intervention
Number	12
Location	Imperial College
Country/Territory	United Kingdom
City	London
Period	20/09/2009 → 24/09/2009

Series	Lecture Notes in Computer Science
ISSN	0302-9743

Keywords

Alignment
Cost functions
Cost reduction
Image registration
Medical computing

Access to Document

10.1007/978-3-642-04268-3_70

OpenUrl availability

Full text

Cite this

@inproceedings{6092723bdaa5449bbad75c9ffe7dbc28,

title = "Asymmetric image-template registration",

abstract = "A natural requirement in pairwise image registration is that the resulting deformation is independent of the order of the images. This constraint is typically achieved via a symmetric cost function and has been shown to reduce the effects of local optima. Consequently, symmetric registration has been successfully applied to pairwise image registration as well as the spatial alignment of individual images with a template. However, recent work has shown that the relationship between an image and a template is fundamentally asymmetric. In this paper, we develop a method that reconciles the practical advantages of symmetric registration with the asymmetric nature of image-template registration by adding a simple correction factor to the symmetric cost function. We instantiate our model within a log-domain diffeomorphic registration framework. Our experiments show exploiting the asymmetry in image-template registration improves alignment in the image coordinates. {\textcopyright} 2009 Springer-Verlag.",

keywords = "Alignment, Cost functions, Cost reduction, Image registration, Medical computing",

author = "Sabuncu, {Mert R.} and Yeo, {B. T. Thomas} and {Van Leemput}, Koen and Tom Vercauteren and Polina Golland",

year = "2009",

doi = "10.1007/978-3-642-04268-3_70",

language = "English",

isbn = "978-3-642-04267-6",

volume = "5761",

series = "Lecture Notes in Computer Science",

publisher = "Springer-verlag Berlin",

pages = "565--573",

booktitle = "Medical Image Computing and Computer-Assisted Intervention",

note = "12th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI2009 ; Conference date: 20-09-2009 Through 24-09-2009",

}

Sabuncu, MR, Yeo, BTT, Van Leemput, K, Vercauteren, T & Golland, P 2009, Asymmetric image-template registration. in Medical Image Computing and Computer-Assisted Intervention. vol. 5761, Springer-verlag Berlin, Lecture Notes in Computer Science, pp. 565-573, 12th International Conference on Medical Image Computing and Computer-Assisted Intervention, London, United Kingdom, 20/09/2009. https://doi.org/10.1007/978-3-642-04268-3_70

TY - GEN

T1 - Asymmetric image-template registration

AU - Sabuncu, Mert R.

AU - Yeo, B. T. Thomas

AU - Van Leemput, Koen

AU - Vercauteren, Tom

AU - Golland, Polina

N1 - Conference code: 12

PY - 2009

Y1 - 2009

N2 - A natural requirement in pairwise image registration is that the resulting deformation is independent of the order of the images. This constraint is typically achieved via a symmetric cost function and has been shown to reduce the effects of local optima. Consequently, symmetric registration has been successfully applied to pairwise image registration as well as the spatial alignment of individual images with a template. However, recent work has shown that the relationship between an image and a template is fundamentally asymmetric. In this paper, we develop a method that reconciles the practical advantages of symmetric registration with the asymmetric nature of image-template registration by adding a simple correction factor to the symmetric cost function. We instantiate our model within a log-domain diffeomorphic registration framework. Our experiments show exploiting the asymmetry in image-template registration improves alignment in the image coordinates. © 2009 Springer-Verlag.

AB - A natural requirement in pairwise image registration is that the resulting deformation is independent of the order of the images. This constraint is typically achieved via a symmetric cost function and has been shown to reduce the effects of local optima. Consequently, symmetric registration has been successfully applied to pairwise image registration as well as the spatial alignment of individual images with a template. However, recent work has shown that the relationship between an image and a template is fundamentally asymmetric. In this paper, we develop a method that reconciles the practical advantages of symmetric registration with the asymmetric nature of image-template registration by adding a simple correction factor to the symmetric cost function. We instantiate our model within a log-domain diffeomorphic registration framework. Our experiments show exploiting the asymmetry in image-template registration improves alignment in the image coordinates. © 2009 Springer-Verlag.

KW - Alignment

KW - Cost functions

KW - Cost reduction

KW - Image registration

KW - Medical computing

U2 - 10.1007/978-3-642-04268-3_70

DO - 10.1007/978-3-642-04268-3_70

M3 - Article in proceedings

SN - 978-3-642-04267-6

VL - 5761

T3 - Lecture Notes in Computer Science

SP - 565

EP - 573

BT - Medical Image Computing and Computer-Assisted Intervention

PB - Springer-verlag Berlin

T2 - 12th International Conference on Medical Image Computing and Computer-Assisted Intervention

Y2 - 20 September 2009 through 24 September 2009

ER -

Asymmetric image-template registration

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this