Deformable image registration using a cue-aware deep regression network

Xiaohuan Cao, Jianhua Yang, Jun Zhang, Qian Wang, Pew Thian Yap, Dinggang Shen

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Significance: Analysis of modern large-scale, multicenter or diseased data requires deformable registration algorithms that can cope with data of diverse nature. Objective: We propose a novel deformable registration method, which is based on a cue-aware deep regression network, to deal with multiple databases with minimal parameter tuning. Methods: Our method learns and predicts the deformation field between a reference image and a subject image. Specifically, given a set of training images, our method learns the displacement vector associated with a pair of reference-subject patches. To achieve this, we first introduce a key-point truncated-balanced sampling strategy to facilitate accurate learning from the image database of limited size. Then, we design a cue-aware deep regression network, where we propose to employ the contextual cue, i.e., the scale-adaptive local similarity, to more apparently guide the learning process. The deep regression network is aware of the contextual cue for accurate prediction of local deformation. Results and Conclusion: Our experiments show that the proposed method can tackle various registration tasks on different databases, giving consistent good performance without the need of manual parameter tuning, which could be applicable to various clinical applications.

Original languageEnglish
Article number8331111
Pages (from-to)1900-1911
Number of pages12
JournalIEEE Transactions on Biomedical Engineering
Volume65
Issue number9
DOIs
Publication statusPublished - 2018 Sep 1

Fingerprint

Image registration
Cues
Tuning
Databases
Learning
Sampling
Experiments

Keywords

  • Deep learning
  • Deformable registration
  • Key-points sampling
  • Nonlinear regression

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Deformable image registration using a cue-aware deep regression network. / Cao, Xiaohuan; Yang, Jianhua; Zhang, Jun; Wang, Qian; Yap, Pew Thian; Shen, Dinggang.

In: IEEE Transactions on Biomedical Engineering, Vol. 65, No. 9, 8331111, 01.09.2018, p. 1900-1911.

Research output: Contribution to journalArticle

Cao, Xiaohuan ; Yang, Jianhua ; Zhang, Jun ; Wang, Qian ; Yap, Pew Thian ; Shen, Dinggang. / Deformable image registration using a cue-aware deep regression network. In: IEEE Transactions on Biomedical Engineering. 2018 ; Vol. 65, No. 9. pp. 1900-1911.
@article{6ee06d70b2ed4051b6237325b8b57fde,
title = "Deformable image registration using a cue-aware deep regression network",
abstract = "Significance: Analysis of modern large-scale, multicenter or diseased data requires deformable registration algorithms that can cope with data of diverse nature. Objective: We propose a novel deformable registration method, which is based on a cue-aware deep regression network, to deal with multiple databases with minimal parameter tuning. Methods: Our method learns and predicts the deformation field between a reference image and a subject image. Specifically, given a set of training images, our method learns the displacement vector associated with a pair of reference-subject patches. To achieve this, we first introduce a key-point truncated-balanced sampling strategy to facilitate accurate learning from the image database of limited size. Then, we design a cue-aware deep regression network, where we propose to employ the contextual cue, i.e., the scale-adaptive local similarity, to more apparently guide the learning process. The deep regression network is aware of the contextual cue for accurate prediction of local deformation. Results and Conclusion: Our experiments show that the proposed method can tackle various registration tasks on different databases, giving consistent good performance without the need of manual parameter tuning, which could be applicable to various clinical applications.",
keywords = "Deep learning, Deformable registration, Key-points sampling, Nonlinear regression",
author = "Xiaohuan Cao and Jianhua Yang and Jun Zhang and Qian Wang and Yap, {Pew Thian} and Dinggang Shen",
year = "2018",
month = "9",
day = "1",
doi = "10.1109/TBME.2018.2822826",
language = "English",
volume = "65",
pages = "1900--1911",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "9",

}

TY - JOUR

T1 - Deformable image registration using a cue-aware deep regression network

AU - Cao, Xiaohuan

AU - Yang, Jianhua

AU - Zhang, Jun

AU - Wang, Qian

AU - Yap, Pew Thian

AU - Shen, Dinggang

PY - 2018/9/1

Y1 - 2018/9/1

N2 - Significance: Analysis of modern large-scale, multicenter or diseased data requires deformable registration algorithms that can cope with data of diverse nature. Objective: We propose a novel deformable registration method, which is based on a cue-aware deep regression network, to deal with multiple databases with minimal parameter tuning. Methods: Our method learns and predicts the deformation field between a reference image and a subject image. Specifically, given a set of training images, our method learns the displacement vector associated with a pair of reference-subject patches. To achieve this, we first introduce a key-point truncated-balanced sampling strategy to facilitate accurate learning from the image database of limited size. Then, we design a cue-aware deep regression network, where we propose to employ the contextual cue, i.e., the scale-adaptive local similarity, to more apparently guide the learning process. The deep regression network is aware of the contextual cue for accurate prediction of local deformation. Results and Conclusion: Our experiments show that the proposed method can tackle various registration tasks on different databases, giving consistent good performance without the need of manual parameter tuning, which could be applicable to various clinical applications.

AB - Significance: Analysis of modern large-scale, multicenter or diseased data requires deformable registration algorithms that can cope with data of diverse nature. Objective: We propose a novel deformable registration method, which is based on a cue-aware deep regression network, to deal with multiple databases with minimal parameter tuning. Methods: Our method learns and predicts the deformation field between a reference image and a subject image. Specifically, given a set of training images, our method learns the displacement vector associated with a pair of reference-subject patches. To achieve this, we first introduce a key-point truncated-balanced sampling strategy to facilitate accurate learning from the image database of limited size. Then, we design a cue-aware deep regression network, where we propose to employ the contextual cue, i.e., the scale-adaptive local similarity, to more apparently guide the learning process. The deep regression network is aware of the contextual cue for accurate prediction of local deformation. Results and Conclusion: Our experiments show that the proposed method can tackle various registration tasks on different databases, giving consistent good performance without the need of manual parameter tuning, which could be applicable to various clinical applications.

KW - Deep learning

KW - Deformable registration

KW - Key-points sampling

KW - Nonlinear regression

UR - http://www.scopus.com/inward/record.url?scp=85052623089&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052623089&partnerID=8YFLogxK

U2 - 10.1109/TBME.2018.2822826

DO - 10.1109/TBME.2018.2822826

M3 - Article

VL - 65

SP - 1900

EP - 1911

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 9

M1 - 8331111

ER -