Deformable image registration based on similarity-steered CNN regression

Xiaohuan Cao, Jianhua Yang, Jun Zhang, Dong Nie, Minjeong Kim, Qian Wang, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

20 Citations (Scopus)

Abstract

Existing deformable registration methods require exhaustively iterative optimization, along with careful parameter tuning, to estimate the deformation field between images. Although some learning-based methods have been proposed for initiating deformation estimation, they are often template-specific and not flexible in practical use. In this paper, we propose a convolutional neural network (CNN) based regression model to directly learn the complex mapping from the input image pair (i.e., a pair of template and subject) to their corresponding deformation field. Specifically, our CNN architecture is designed in a patch-based manner to learn the complex mapping from the input patch pairs to their respective deformation field. First, the equalized active-points guided sampling strategy is introduced to facilitate accurate CNN model learning upon a limited image dataset. Then, the similarity-steered CNN architecture is designed, where we propose to add the auxiliary contextual cue, i.e., the similarity between input patches, to more directly guide the learning process. Experiments on different brain image datasets demonstrate promising registration performance based on our CNN model. Furthermore, it is found that the trained CNN model from one dataset can be successfully transferred to another dataset, although brain appearances across datasets are quite variable.

Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings
PublisherSpringer Verlag
Pages300-308
Number of pages9
Volume10433 LNCS
ISBN (Print)9783319661810
DOIs
Publication statusPublished - 2017
Externally publishedYes
Event20th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2017 - Quebec City, Canada
Duration: 2017 Sep 112017 Sep 13

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10433 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other20th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2017
CountryCanada
CityQuebec City
Period17/9/1117/9/13

Fingerprint

Image registration
Image Registration
Regression
Neural Network Model
Neural Networks
Neural networks
Patch
Network Architecture
Registration
Template
Network architecture
Sampling Strategy
Brain
Parameter Tuning
Learning Process
Regression Model
Similarity
Optimization
Tuning
Estimate

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Cao, X., Yang, J., Zhang, J., Nie, D., Kim, M., Wang, Q., & Shen, D. (2017). Deformable image registration based on similarity-steered CNN regression. In Medical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings (Vol. 10433 LNCS, pp. 300-308). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10433 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-66182-7_35

Deformable image registration based on similarity-steered CNN regression. / Cao, Xiaohuan; Yang, Jianhua; Zhang, Jun; Nie, Dong; Kim, Minjeong; Wang, Qian; Shen, Dinggang.

Medical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings. Vol. 10433 LNCS Springer Verlag, 2017. p. 300-308 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10433 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Cao, X, Yang, J, Zhang, J, Nie, D, Kim, M, Wang, Q & Shen, D 2017, Deformable image registration based on similarity-steered CNN regression. in Medical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings. vol. 10433 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10433 LNCS, Springer Verlag, pp. 300-308, 20th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2017, Quebec City, Canada, 17/9/11. https://doi.org/10.1007/978-3-319-66182-7_35
Cao X, Yang J, Zhang J, Nie D, Kim M, Wang Q et al. Deformable image registration based on similarity-steered CNN regression. In Medical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings. Vol. 10433 LNCS. Springer Verlag. 2017. p. 300-308. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-66182-7_35
Cao, Xiaohuan ; Yang, Jianhua ; Zhang, Jun ; Nie, Dong ; Kim, Minjeong ; Wang, Qian ; Shen, Dinggang. / Deformable image registration based on similarity-steered CNN regression. Medical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings. Vol. 10433 LNCS Springer Verlag, 2017. pp. 300-308 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{7bf52100479f4b43b30fe4faf887e9e7,
title = "Deformable image registration based on similarity-steered CNN regression",
abstract = "Existing deformable registration methods require exhaustively iterative optimization, along with careful parameter tuning, to estimate the deformation field between images. Although some learning-based methods have been proposed for initiating deformation estimation, they are often template-specific and not flexible in practical use. In this paper, we propose a convolutional neural network (CNN) based regression model to directly learn the complex mapping from the input image pair (i.e., a pair of template and subject) to their corresponding deformation field. Specifically, our CNN architecture is designed in a patch-based manner to learn the complex mapping from the input patch pairs to their respective deformation field. First, the equalized active-points guided sampling strategy is introduced to facilitate accurate CNN model learning upon a limited image dataset. Then, the similarity-steered CNN architecture is designed, where we propose to add the auxiliary contextual cue, i.e., the similarity between input patches, to more directly guide the learning process. Experiments on different brain image datasets demonstrate promising registration performance based on our CNN model. Furthermore, it is found that the trained CNN model from one dataset can be successfully transferred to another dataset, although brain appearances across datasets are quite variable.",
author = "Xiaohuan Cao and Jianhua Yang and Jun Zhang and Dong Nie and Minjeong Kim and Qian Wang and Dinggang Shen",
year = "2017",
doi = "10.1007/978-3-319-66182-7_35",
language = "English",
isbn = "9783319661810",
volume = "10433 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "300--308",
booktitle = "Medical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings",

}

TY - GEN

T1 - Deformable image registration based on similarity-steered CNN regression

AU - Cao, Xiaohuan

AU - Yang, Jianhua

AU - Zhang, Jun

AU - Nie, Dong

AU - Kim, Minjeong

AU - Wang, Qian

AU - Shen, Dinggang

PY - 2017

Y1 - 2017

N2 - Existing deformable registration methods require exhaustively iterative optimization, along with careful parameter tuning, to estimate the deformation field between images. Although some learning-based methods have been proposed for initiating deformation estimation, they are often template-specific and not flexible in practical use. In this paper, we propose a convolutional neural network (CNN) based regression model to directly learn the complex mapping from the input image pair (i.e., a pair of template and subject) to their corresponding deformation field. Specifically, our CNN architecture is designed in a patch-based manner to learn the complex mapping from the input patch pairs to their respective deformation field. First, the equalized active-points guided sampling strategy is introduced to facilitate accurate CNN model learning upon a limited image dataset. Then, the similarity-steered CNN architecture is designed, where we propose to add the auxiliary contextual cue, i.e., the similarity between input patches, to more directly guide the learning process. Experiments on different brain image datasets demonstrate promising registration performance based on our CNN model. Furthermore, it is found that the trained CNN model from one dataset can be successfully transferred to another dataset, although brain appearances across datasets are quite variable.

AB - Existing deformable registration methods require exhaustively iterative optimization, along with careful parameter tuning, to estimate the deformation field between images. Although some learning-based methods have been proposed for initiating deformation estimation, they are often template-specific and not flexible in practical use. In this paper, we propose a convolutional neural network (CNN) based regression model to directly learn the complex mapping from the input image pair (i.e., a pair of template and subject) to their corresponding deformation field. Specifically, our CNN architecture is designed in a patch-based manner to learn the complex mapping from the input patch pairs to their respective deformation field. First, the equalized active-points guided sampling strategy is introduced to facilitate accurate CNN model learning upon a limited image dataset. Then, the similarity-steered CNN architecture is designed, where we propose to add the auxiliary contextual cue, i.e., the similarity between input patches, to more directly guide the learning process. Experiments on different brain image datasets demonstrate promising registration performance based on our CNN model. Furthermore, it is found that the trained CNN model from one dataset can be successfully transferred to another dataset, although brain appearances across datasets are quite variable.

UR - http://www.scopus.com/inward/record.url?scp=85029377408&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029377408&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-66182-7_35

DO - 10.1007/978-3-319-66182-7_35

M3 - Conference contribution

AN - SCOPUS:85029377408

SN - 9783319661810

VL - 10433 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 300

EP - 308

BT - Medical Image Computing and Computer Assisted Intervention − MICCAI 2017 - 20th International Conference, Proceedings

PB - Springer Verlag

ER -