Diffeomorphic image registration with automatic time-step adjustment

Akshay Sadananda Uppinakudru Pai; S. Klein; Stefan Horst Sommer; Sune Darkner; Jon Sporring; Mads Nielsen

doi:10.1109/ISBI.2015.7164060

Diffeomorphic image registration with automatic time-step adjustment

Akshay Sadananda Uppinakudru Pai, S. Klein, Stefan Horst Sommer, Sune Darkner, Jon Sporring, Mads Nielsen

Department of Computer Science

1 Citation (Scopus)

Abstract

In this paper, we propose an automated Euler's time-step adjustment scheme for diffeomorphic image registration using stationary velocity fields (SVFs). The proposed variational problem aims at bounding the inverse consistency error by adaptively adjusting the number of Euler's step required to realize the time integration. This particular formulation allows us to gain computationally since only relevant number of time steps are taken. We parameterize the SVFs using multi-scale Wendland kernels through the kernel bundle framework. In terms of performance, the proposed scheme reaches the same accuracy as a fixed time-step scheme however at a much less computational cost.

Original language	English
Title of host publication	Biomedical Imaging (ISBI), 2015 IEEE 12th International Symposium on
Number of pages	4
Publisher	IEEE
Publication date	21 Jul 2015
Pages	1085-1088
DOIs	https://doi.org/10.1109/ISBI.2015.7164060
Publication status	Published - 21 Jul 2015

Access to Document

10.1109/ISBI.2015.7164060

Cite this

@inbook{fcfd5657fe96471e94550075be09bb19,

title = "Diffeomorphic image registration with automatic time-step adjustment",

abstract = "In this paper, we propose an automated Euler's time-step adjustment scheme for diffeomorphic image registration using stationary velocity fields (SVFs). The proposed variational problem aims at bounding the inverse consistency error by adaptively adjusting the number of Euler's step required to realize the time integration. This particular formulation allows us to gain computationally since only relevant number of time steps are taken. We parameterize the SVFs using multi-scale Wendland kernels through the kernel bundle framework. In terms of performance, the proposed scheme reaches the same accuracy as a fixed time-step scheme however at a much less computational cost.",

keywords = "adaptive systems, image registration, medical image processing, operating system kernels, variational techniques, SVF parameterization, adaptive Euler step number adjustment, automated Euler time-step adjustment, automatic time-step adjustment, computational cost reduction, diffeomorphic image registration, fixed time-step scheme, inverse consistency error bounding, kernel bundle framework, multiscale Wendland kernel, stationary velocity field, time integration, variational problem, Accuracy, Biomedical imaging, Cost function, Erbium, Image registration, Indexes, Kernel, Diffeomorphic Registration, Euler's scheme, Inverse consistency, Kernel Bundle, Wendland Kernels",

author = "Pai, {Akshay Sadananda Uppinakudru} and S. Klein and Sommer, {Stefan Horst} and Sune Darkner and Jon Sporring and Mads Nielsen",

year = "2015",

month = jul,

day = "21",

doi = "10.1109/ISBI.2015.7164060",

language = "English",

pages = "1085--1088",

booktitle = "Biomedical Imaging (ISBI), 2015 IEEE 12th International Symposium on",

publisher = "IEEE",

}

TY - CHAP

T1 - Diffeomorphic image registration with automatic time-step adjustment

AU - Pai, Akshay Sadananda Uppinakudru

AU - Klein, S.

AU - Sommer, Stefan Horst

AU - Darkner, Sune

AU - Sporring, Jon

AU - Nielsen, Mads

PY - 2015/7/21

Y1 - 2015/7/21

N2 - In this paper, we propose an automated Euler's time-step adjustment scheme for diffeomorphic image registration using stationary velocity fields (SVFs). The proposed variational problem aims at bounding the inverse consistency error by adaptively adjusting the number of Euler's step required to realize the time integration. This particular formulation allows us to gain computationally since only relevant number of time steps are taken. We parameterize the SVFs using multi-scale Wendland kernels through the kernel bundle framework. In terms of performance, the proposed scheme reaches the same accuracy as a fixed time-step scheme however at a much less computational cost.

AB - In this paper, we propose an automated Euler's time-step adjustment scheme for diffeomorphic image registration using stationary velocity fields (SVFs). The proposed variational problem aims at bounding the inverse consistency error by adaptively adjusting the number of Euler's step required to realize the time integration. This particular formulation allows us to gain computationally since only relevant number of time steps are taken. We parameterize the SVFs using multi-scale Wendland kernels through the kernel bundle framework. In terms of performance, the proposed scheme reaches the same accuracy as a fixed time-step scheme however at a much less computational cost.

KW - adaptive systems

KW - image registration

KW - medical image processing

KW - operating system kernels

KW - variational techniques

KW - SVF parameterization

KW - adaptive Euler step number adjustment

KW - automated Euler time-step adjustment

KW - automatic time-step adjustment

KW - computational cost reduction

KW - diffeomorphic image registration

KW - fixed time-step scheme

KW - inverse consistency error bounding

KW - kernel bundle framework

KW - multiscale Wendland kernel

KW - stationary velocity field

KW - time integration

KW - variational problem

KW - Accuracy

KW - Biomedical imaging

KW - Cost function

KW - Erbium

KW - Image registration

KW - Indexes

KW - Kernel

KW - Diffeomorphic Registration

KW - Euler's scheme

KW - Inverse consistency

KW - Kernel Bundle

KW - Wendland Kernels

U2 - 10.1109/ISBI.2015.7164060

DO - 10.1109/ISBI.2015.7164060

M3 - Book chapter

SP - 1085

EP - 1088

BT - Biomedical Imaging (ISBI), 2015 IEEE 12th International Symposium on

PB - IEEE

ER -