Sparsity and scale: compact representations of deformation for diffeomorphic registration

Stefan Horst Sommer; Mads Nielsen; Xavier Pennec

doi:10.1109/MMBIA.2012.6164753

Sparsity and scale: compact representations of deformation for diffeomorphic registration

Stefan Horst Sommer, Mads Nielsen, Xavier Pennec

Department of Computer Science

2 Citations (Scopus)

Abstract

In order to detect small-scale deformations during disease propagation while allowing large-scale deformation needed for inter-subject registration, we wish to model deformation at multiple scales and represent the deformation at the relevant scales only. With the LDDMM registration framework, enforcing sparsity results in compact representations but with limited ability to represent deformation across scales. In contrast, the LDDKBM extension of LDDMM allows representations of deformation at multiple scales but it does not favour compactness and hence may represent deformation at more scales than necessary. In this paper, we combine a sparsity prior with the multi-scale framework resulting in an algorithm allowing compact representation of deformation across scales. We present a mathematical formulation of the algorithm and evaluate it on a dataset of annotated lung CT images.

Original language	English
Title of host publication	2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA)
Number of pages	8
Publisher	IEEE
Publication date	2012
Pages	217-224
ISBN (Print)	978-1-4673-0352-1
ISBN (Electronic)	978-1-4673-0353-8
DOIs	https://doi.org/10.1109/MMBIA.2012.6164753
Publication status	Published - 2012
Event	2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA) - Breckenridge, CO, United States Duration: 9 Jan 2012 → 10 Jan 2012

Workshop

Workshop	2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA)
Country/Territory	United States
City	Breckenridge, CO
Period	09/01/2012 → 10/01/2012

Access to Document

10.1109/MMBIA.2012.6164753

Cite this

Sommer, SH , Nielsen, M & Pennec, X 2012, Sparsity and scale: compact representations of deformation for diffeomorphic registration. in 2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA). IEEE, pp. 217-224 , 2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA), Breckenridge, CO, United States, 09/01/2012. https://doi.org/10.1109/MMBIA.2012.6164753

@inproceedings{e5a87c6be66040ccad751e291ca830ee,

title = "Sparsity and scale: compact representations of deformation for diffeomorphic registration",

abstract = "In order to detect small-scale deformations during disease propagation while allowing large-scale deformation needed for inter-subject registration, we wish to model deformation at multiple scales and represent the deformation at the relevant scales only. With the LDDMM registration framework, enforcing sparsity results in compact representations but with limited ability to represent deformation across scales. In contrast, the LDDKBM extension of LDDMM allows representations of deformation at multiple scales but it does not favour compactness and hence may represent deformation at more scales than necessary. In this paper, we combine a sparsity prior with the multi-scale framework resulting in an algorithm allowing compact representation of deformation across scales. We present a mathematical formulation of the algorithm and evaluate it on a dataset of annotated lung CT images.",

author = "Sommer, {Stefan Horst} and Mads Nielsen and Xavier Pennec",

year = "2012",

doi = "10.1109/MMBIA.2012.6164753",

language = "English",

isbn = "978-1-4673-0352-1",

pages = "217--224 ",

booktitle = "2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA)",

publisher = "IEEE",

note = "2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA) ; Conference date: 09-01-2012 Through 10-01-2012",

}

TY - GEN

T1 - Sparsity and scale

T2 - 2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA)

AU - Sommer, Stefan Horst

AU - Nielsen, Mads

AU - Pennec, Xavier

PY - 2012

Y1 - 2012

N2 - In order to detect small-scale deformations during disease propagation while allowing large-scale deformation needed for inter-subject registration, we wish to model deformation at multiple scales and represent the deformation at the relevant scales only. With the LDDMM registration framework, enforcing sparsity results in compact representations but with limited ability to represent deformation across scales. In contrast, the LDDKBM extension of LDDMM allows representations of deformation at multiple scales but it does not favour compactness and hence may represent deformation at more scales than necessary. In this paper, we combine a sparsity prior with the multi-scale framework resulting in an algorithm allowing compact representation of deformation across scales. We present a mathematical formulation of the algorithm and evaluate it on a dataset of annotated lung CT images.

AB - In order to detect small-scale deformations during disease propagation while allowing large-scale deformation needed for inter-subject registration, we wish to model deformation at multiple scales and represent the deformation at the relevant scales only. With the LDDMM registration framework, enforcing sparsity results in compact representations but with limited ability to represent deformation across scales. In contrast, the LDDKBM extension of LDDMM allows representations of deformation at multiple scales but it does not favour compactness and hence may represent deformation at more scales than necessary. In this paper, we combine a sparsity prior with the multi-scale framework resulting in an algorithm allowing compact representation of deformation across scales. We present a mathematical formulation of the algorithm and evaluate it on a dataset of annotated lung CT images.

U2 - 10.1109/MMBIA.2012.6164753

DO - 10.1109/MMBIA.2012.6164753

M3 - Article in proceedings

SN - 978-1-4673-0352-1

SP - 217

EP - 224

BT - 2012 IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA)

PB - IEEE

Y2 - 9 January 2012 through 10 January 2012

ER -