A dynamic tree-based registration could handle possible large deformations among MR brain images

Pei Zhang, Guorong Wu, Yaozong Gao, Pew Thian Yap, Dinggang Shen

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Multi-atlas segmentation is a powerful approach to automated anatomy delineation via fusing label information from a set of spatially normalized atlases. For simplicity, many existing methods perform pairwise image registration, leading to inaccurate segmentation especially when shape variation is large. In this paper, we propose a dynamic tree-based strategy for effective large-deformation registration and multi-atlas segmentation. To deal with local minima caused by large shape variation, coarse estimates of deformations are first obtained via alignment of automatically localized landmark points. The dynamic tree capturing the structural relationships between images is then employed to further reduce misalignment errors. Evaluation based on two real human brain datasets, ADNI and LPBA40, shows that our method significantly improves registration and segmentation accuracy.

Original languageEnglish
Pages (from-to)1-7
Number of pages7
JournalComputerized Medical Imaging and Graphics
Volume52
DOIs
Publication statusPublished - 2016 Sep 1

Fingerprint

Atlases
Brain
Image registration
Labels
Anatomy

Keywords

  • Corresponding points
  • Dynamic tree
  • Large-deformation image registration
  • Multi-atlas segmentation

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Health Informatics
  • Radiological and Ultrasound Technology
  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Cite this

A dynamic tree-based registration could handle possible large deformations among MR brain images. / Zhang, Pei; Wu, Guorong; Gao, Yaozong; Yap, Pew Thian; Shen, Dinggang.

In: Computerized Medical Imaging and Graphics, Vol. 52, 01.09.2016, p. 1-7.

Research output: Contribution to journalArticle

@article{106936a23d5f45b288f40c8debf40981,
title = "A dynamic tree-based registration could handle possible large deformations among MR brain images",
abstract = "Multi-atlas segmentation is a powerful approach to automated anatomy delineation via fusing label information from a set of spatially normalized atlases. For simplicity, many existing methods perform pairwise image registration, leading to inaccurate segmentation especially when shape variation is large. In this paper, we propose a dynamic tree-based strategy for effective large-deformation registration and multi-atlas segmentation. To deal with local minima caused by large shape variation, coarse estimates of deformations are first obtained via alignment of automatically localized landmark points. The dynamic tree capturing the structural relationships between images is then employed to further reduce misalignment errors. Evaluation based on two real human brain datasets, ADNI and LPBA40, shows that our method significantly improves registration and segmentation accuracy.",
keywords = "Corresponding points, Dynamic tree, Large-deformation image registration, Multi-atlas segmentation",
author = "Pei Zhang and Guorong Wu and Yaozong Gao and Yap, {Pew Thian} and Dinggang Shen",
year = "2016",
month = "9",
day = "1",
doi = "10.1016/j.compmedimag.2016.04.005",
language = "English",
volume = "52",
pages = "1--7",
journal = "Computerized Medical Imaging and Graphics",
issn = "0895-6111",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - A dynamic tree-based registration could handle possible large deformations among MR brain images

AU - Zhang, Pei

AU - Wu, Guorong

AU - Gao, Yaozong

AU - Yap, Pew Thian

AU - Shen, Dinggang

PY - 2016/9/1

Y1 - 2016/9/1

N2 - Multi-atlas segmentation is a powerful approach to automated anatomy delineation via fusing label information from a set of spatially normalized atlases. For simplicity, many existing methods perform pairwise image registration, leading to inaccurate segmentation especially when shape variation is large. In this paper, we propose a dynamic tree-based strategy for effective large-deformation registration and multi-atlas segmentation. To deal with local minima caused by large shape variation, coarse estimates of deformations are first obtained via alignment of automatically localized landmark points. The dynamic tree capturing the structural relationships between images is then employed to further reduce misalignment errors. Evaluation based on two real human brain datasets, ADNI and LPBA40, shows that our method significantly improves registration and segmentation accuracy.

AB - Multi-atlas segmentation is a powerful approach to automated anatomy delineation via fusing label information from a set of spatially normalized atlases. For simplicity, many existing methods perform pairwise image registration, leading to inaccurate segmentation especially when shape variation is large. In this paper, we propose a dynamic tree-based strategy for effective large-deformation registration and multi-atlas segmentation. To deal with local minima caused by large shape variation, coarse estimates of deformations are first obtained via alignment of automatically localized landmark points. The dynamic tree capturing the structural relationships between images is then employed to further reduce misalignment errors. Evaluation based on two real human brain datasets, ADNI and LPBA40, shows that our method significantly improves registration and segmentation accuracy.

KW - Corresponding points

KW - Dynamic tree

KW - Large-deformation image registration

KW - Multi-atlas segmentation

UR - http://www.scopus.com/inward/record.url?scp=84971665297&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84971665297&partnerID=8YFLogxK

U2 - 10.1016/j.compmedimag.2016.04.005

DO - 10.1016/j.compmedimag.2016.04.005

M3 - Article

C2 - 27235894

AN - SCOPUS:84971665297

VL - 52

SP - 1

EP - 7

JO - Computerized Medical Imaging and Graphics

JF - Computerized Medical Imaging and Graphics

SN - 0895-6111

ER -