Longitudinal Image Analysis via Path Regression on the Image Manifold

Shi Hui Ying, Xiao Fang Zhang, Ya Xin Peng, Dinggang Shen

Research output: Contribution to journalArticle

Abstract

Longitudinal image analysis plays an important role in depicting the development of the brain structure, where image regression and interpolation are two commonly used techniques. In this paper, we develop an efficient model and approach based on a path regression on the image manifold instead of the geodesic regression to avoid the complexity of the geodesic computation. Concretely, first we model the deformation by diffeomorphism; then, a large deformation is represented by a path on the orbit of the diffeomorphism group action. This path is obtained by compositing several small deformations, which can be well approximated by its linearization. Second, we introduce some intermediate images as constraints to the model, which guides to form the best-fitting path. Thirdly, we propose an approximated quadratic model by local linearization method, where a closed form is deduced for the solution. It actually speeds up the algorithm. Finally, we evaluate the proposed model and algorithm on a synthetic data and a real longitudinal MRI data. The results show that our proposed method outperforms several state-of-the-art methods.

Original languageEnglish
JournalJournal of the Operations Research Society of China
DOIs
Publication statusPublished - 2019 Jan 1
Externally publishedYes

Fingerprint

Image analysis
Linearization
Interpolation

Keywords

  • Diffeomorphism group
  • Image registration
  • Infant brain development
  • Longitudinal image analysis
  • Path regression

ASJC Scopus subject areas

  • Decision Sciences(all)

Cite this

Longitudinal Image Analysis via Path Regression on the Image Manifold. / Ying, Shi Hui; Zhang, Xiao Fang; Peng, Ya Xin; Shen, Dinggang.

In: Journal of the Operations Research Society of China, 01.01.2019.

Research output: Contribution to journalArticle

@article{41f0e522baec4fb8b7aaad2fbf073aa5,
title = "Longitudinal Image Analysis via Path Regression on the Image Manifold",
abstract = "Longitudinal image analysis plays an important role in depicting the development of the brain structure, where image regression and interpolation are two commonly used techniques. In this paper, we develop an efficient model and approach based on a path regression on the image manifold instead of the geodesic regression to avoid the complexity of the geodesic computation. Concretely, first we model the deformation by diffeomorphism; then, a large deformation is represented by a path on the orbit of the diffeomorphism group action. This path is obtained by compositing several small deformations, which can be well approximated by its linearization. Second, we introduce some intermediate images as constraints to the model, which guides to form the best-fitting path. Thirdly, we propose an approximated quadratic model by local linearization method, where a closed form is deduced for the solution. It actually speeds up the algorithm. Finally, we evaluate the proposed model and algorithm on a synthetic data and a real longitudinal MRI data. The results show that our proposed method outperforms several state-of-the-art methods.",
keywords = "Diffeomorphism group, Image registration, Infant brain development, Longitudinal image analysis, Path regression",
author = "Ying, {Shi Hui} and Zhang, {Xiao Fang} and Peng, {Ya Xin} and Dinggang Shen",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s40305-019-00251-2",
language = "English",
journal = "Journal of the Operations Research Society of China",
issn = "2194-668X",
publisher = "Springer Science + Business Media",

}

TY - JOUR

T1 - Longitudinal Image Analysis via Path Regression on the Image Manifold

AU - Ying, Shi Hui

AU - Zhang, Xiao Fang

AU - Peng, Ya Xin

AU - Shen, Dinggang

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Longitudinal image analysis plays an important role in depicting the development of the brain structure, where image regression and interpolation are two commonly used techniques. In this paper, we develop an efficient model and approach based on a path regression on the image manifold instead of the geodesic regression to avoid the complexity of the geodesic computation. Concretely, first we model the deformation by diffeomorphism; then, a large deformation is represented by a path on the orbit of the diffeomorphism group action. This path is obtained by compositing several small deformations, which can be well approximated by its linearization. Second, we introduce some intermediate images as constraints to the model, which guides to form the best-fitting path. Thirdly, we propose an approximated quadratic model by local linearization method, where a closed form is deduced for the solution. It actually speeds up the algorithm. Finally, we evaluate the proposed model and algorithm on a synthetic data and a real longitudinal MRI data. The results show that our proposed method outperforms several state-of-the-art methods.

AB - Longitudinal image analysis plays an important role in depicting the development of the brain structure, where image regression and interpolation are two commonly used techniques. In this paper, we develop an efficient model and approach based on a path regression on the image manifold instead of the geodesic regression to avoid the complexity of the geodesic computation. Concretely, first we model the deformation by diffeomorphism; then, a large deformation is represented by a path on the orbit of the diffeomorphism group action. This path is obtained by compositing several small deformations, which can be well approximated by its linearization. Second, we introduce some intermediate images as constraints to the model, which guides to form the best-fitting path. Thirdly, we propose an approximated quadratic model by local linearization method, where a closed form is deduced for the solution. It actually speeds up the algorithm. Finally, we evaluate the proposed model and algorithm on a synthetic data and a real longitudinal MRI data. The results show that our proposed method outperforms several state-of-the-art methods.

KW - Diffeomorphism group

KW - Image registration

KW - Infant brain development

KW - Longitudinal image analysis

KW - Path regression

UR - http://www.scopus.com/inward/record.url?scp=85067229612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067229612&partnerID=8YFLogxK

U2 - 10.1007/s40305-019-00251-2

DO - 10.1007/s40305-019-00251-2

M3 - Article

JO - Journal of the Operations Research Society of China

JF - Journal of the Operations Research Society of China

SN - 2194-668X

ER -