Deep learning deformation initialization for rapid groupwise registration of inhomogeneous image populations

Sahar Ahmad, Jingfan Fan, Pei Dong, Xiaohuan Cao, Pew Thian Yap, Dinggang Shen

Research output: Contribution to journalArticle

Abstract

Groupwise image registration tackles biases that can potentially arise from inappropriate template selection. It typically involves simultaneous registration of a cohort of images to a common space that is not specified a priori. Existing groupwise registration methods are computationally complex and are only effective for image populations without large anatomical variations. In this paper, we propose a deep learning framework to rapidly estimate large deformations between images to significantly reduce structural variability. Specifically, we employ a multi-level graph coarsening method to agglomerate similar images into clusters, each represented by an exemplar image. We then use a deep learning framework to predict the initial deformations between images. Warping with the estimated deformations brings the images closer in the image manifold and their alignment can be further refined using conventional groupwise registration algorithms. We evaluated the effectiveness of our method in groupwise registration of MR brain images and compared it against state-of-the-art groupwise registration methods. Experimental results indicate that deformation initialization enables groupwise registration to converge significantly faster with competitive accuracy, therefore facilitates large-scale imaging studies.

Original languageEnglish
Article number34
JournalFrontiers in Neuroinformatics
Volume13
DOIs
Publication statusPublished - 2019 Apr 16

Fingerprint

Learning
Population
Image registration
Coarsening
Brain
Imaging techniques
Deep learning

Keywords

  • Brain templates
  • Convolutional neural network
  • Deep learning
  • Graph coarsening
  • Groupwise registration
  • MRI

ASJC Scopus subject areas

  • Neuroscience (miscellaneous)
  • Biomedical Engineering
  • Computer Science Applications

Cite this

Deep learning deformation initialization for rapid groupwise registration of inhomogeneous image populations. / Ahmad, Sahar; Fan, Jingfan; Dong, Pei; Cao, Xiaohuan; Yap, Pew Thian; Shen, Dinggang.

In: Frontiers in Neuroinformatics, Vol. 13, 34, 16.04.2019.

Research output: Contribution to journalArticle

@article{73336b38d67548148059e30ef1a8d024,
title = "Deep learning deformation initialization for rapid groupwise registration of inhomogeneous image populations",
abstract = "Groupwise image registration tackles biases that can potentially arise from inappropriate template selection. It typically involves simultaneous registration of a cohort of images to a common space that is not specified a priori. Existing groupwise registration methods are computationally complex and are only effective for image populations without large anatomical variations. In this paper, we propose a deep learning framework to rapidly estimate large deformations between images to significantly reduce structural variability. Specifically, we employ a multi-level graph coarsening method to agglomerate similar images into clusters, each represented by an exemplar image. We then use a deep learning framework to predict the initial deformations between images. Warping with the estimated deformations brings the images closer in the image manifold and their alignment can be further refined using conventional groupwise registration algorithms. We evaluated the effectiveness of our method in groupwise registration of MR brain images and compared it against state-of-the-art groupwise registration methods. Experimental results indicate that deformation initialization enables groupwise registration to converge significantly faster with competitive accuracy, therefore facilitates large-scale imaging studies.",
keywords = "Brain templates, Convolutional neural network, Deep learning, Graph coarsening, Groupwise registration, MRI",
author = "Sahar Ahmad and Jingfan Fan and Pei Dong and Xiaohuan Cao and Yap, {Pew Thian} and Dinggang Shen",
year = "2019",
month = "4",
day = "16",
doi = "10.3389/fninf.2019.00034",
language = "English",
volume = "13",
journal = "Frontiers in Neuroinformatics",
issn = "1662-5196",
publisher = "Frontiers Research Foundation",

}

TY - JOUR

T1 - Deep learning deformation initialization for rapid groupwise registration of inhomogeneous image populations

AU - Ahmad, Sahar

AU - Fan, Jingfan

AU - Dong, Pei

AU - Cao, Xiaohuan

AU - Yap, Pew Thian

AU - Shen, Dinggang

PY - 2019/4/16

Y1 - 2019/4/16

N2 - Groupwise image registration tackles biases that can potentially arise from inappropriate template selection. It typically involves simultaneous registration of a cohort of images to a common space that is not specified a priori. Existing groupwise registration methods are computationally complex and are only effective for image populations without large anatomical variations. In this paper, we propose a deep learning framework to rapidly estimate large deformations between images to significantly reduce structural variability. Specifically, we employ a multi-level graph coarsening method to agglomerate similar images into clusters, each represented by an exemplar image. We then use a deep learning framework to predict the initial deformations between images. Warping with the estimated deformations brings the images closer in the image manifold and their alignment can be further refined using conventional groupwise registration algorithms. We evaluated the effectiveness of our method in groupwise registration of MR brain images and compared it against state-of-the-art groupwise registration methods. Experimental results indicate that deformation initialization enables groupwise registration to converge significantly faster with competitive accuracy, therefore facilitates large-scale imaging studies.

AB - Groupwise image registration tackles biases that can potentially arise from inappropriate template selection. It typically involves simultaneous registration of a cohort of images to a common space that is not specified a priori. Existing groupwise registration methods are computationally complex and are only effective for image populations without large anatomical variations. In this paper, we propose a deep learning framework to rapidly estimate large deformations between images to significantly reduce structural variability. Specifically, we employ a multi-level graph coarsening method to agglomerate similar images into clusters, each represented by an exemplar image. We then use a deep learning framework to predict the initial deformations between images. Warping with the estimated deformations brings the images closer in the image manifold and their alignment can be further refined using conventional groupwise registration algorithms. We evaluated the effectiveness of our method in groupwise registration of MR brain images and compared it against state-of-the-art groupwise registration methods. Experimental results indicate that deformation initialization enables groupwise registration to converge significantly faster with competitive accuracy, therefore facilitates large-scale imaging studies.

KW - Brain templates

KW - Convolutional neural network

KW - Deep learning

KW - Graph coarsening

KW - Groupwise registration

KW - MRI

UR - http://www.scopus.com/inward/record.url?scp=85067406172&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067406172&partnerID=8YFLogxK

U2 - 10.3389/fninf.2019.00034

DO - 10.3389/fninf.2019.00034

M3 - Article

AN - SCOPUS:85067406172

VL - 13

JO - Frontiers in Neuroinformatics

JF - Frontiers in Neuroinformatics

SN - 1662-5196

M1 - 34

ER -