Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration

Dongming Wei, Sahar Ahmad, Zhengwang Wu, Xiaohuan Cao, Xuhua Ren, Gang Li, Dinggang Shen, Qian Wang

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

Abstract

Brain MR image registration is challenging due to the large inter-subject anatomical variation. Especially, the highly convoluted brain cortex makes it difficult to accurately align the corresponding structures of the underlying images. In this paper, we propose a novel deep learning strategy to simplify the image registration task. Specifically, we train a morphological simplification network (MS-Net), which can generate a simplified image with fewer anatomical details given a complex input image. With this trained MS-Net, we can reduce the complexity of both the fixed and the moving images and iteratively derive their respective trajectories of gradually simplified images. The generated images at the ends of the two trajectories are so simple that they are very similar in appearance and morphology and thus easy to register. In this way, these two trajectories can act as a bridge to link the fixed and the moving images and guide their registration. Our experiments show that the proposed method can achieve more accurate registration results than state-of-the-art methods. Moreover, the proposed method can be generalized to the unseen dataset without the need for re-training or domain adaptation.

Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings
EditorsHeung-Il Suk, Mingxia Liu, Chunfeng Lian, Pingkun Yan
PublisherSpringer
Pages203-211
Number of pages9
ISBN (Print)9783030326913
DOIs
Publication statusPublished - 2019 Jan 1
Externally publishedYes
Event10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China
Duration: 2019 Oct 132019 Oct 13

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11861 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
CountryChina
CityShenzhen
Period19/10/1319/10/13

Keywords

  • Brain MRI
  • Deep learning
  • Deformable registration
  • Morphology

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Wei, D., Ahmad, S., Wu, Z., Cao, X., Ren, X., Li, G., ... Wang, Q. (2019). Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration. In H-I. Suk, M. Liu, C. Lian, & P. Yan (Eds.), Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings (pp. 203-211). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11861 LNCS). Springer. https://doi.org/10.1007/978-3-030-32692-0_24

Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration. / Wei, Dongming; Ahmad, Sahar; Wu, Zhengwang; Cao, Xiaohuan; Ren, Xuhua; Li, Gang; Shen, Dinggang; Wang, Qian.

Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. ed. / Heung-Il Suk; Mingxia Liu; Chunfeng Lian; Pingkun Yan. Springer, 2019. p. 203-211 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11861 LNCS).

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

Wei, D, Ahmad, S, Wu, Z, Cao, X, Ren, X, Li, G, Shen, D & Wang, Q 2019, Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration. in H-I Suk, M Liu, C Lian & P Yan (eds), Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11861 LNCS, Springer, pp. 203-211, 10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019, Shenzhen, China, 19/10/13. https://doi.org/10.1007/978-3-030-32692-0_24
Wei D, Ahmad S, Wu Z, Cao X, Ren X, Li G et al. Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration. In Suk H-I, Liu M, Lian C, Yan P, editors, Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. Springer. 2019. p. 203-211. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-32692-0_24
Wei, Dongming ; Ahmad, Sahar ; Wu, Zhengwang ; Cao, Xiaohuan ; Ren, Xuhua ; Li, Gang ; Shen, Dinggang ; Wang, Qian. / Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration. Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. editor / Heung-Il Suk ; Mingxia Liu ; Chunfeng Lian ; Pingkun Yan. Springer, 2019. pp. 203-211 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b190b9304cdf46ee9aa7936e63400509,
title = "Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration",
abstract = "Brain MR image registration is challenging due to the large inter-subject anatomical variation. Especially, the highly convoluted brain cortex makes it difficult to accurately align the corresponding structures of the underlying images. In this paper, we propose a novel deep learning strategy to simplify the image registration task. Specifically, we train a morphological simplification network (MS-Net), which can generate a simplified image with fewer anatomical details given a complex input image. With this trained MS-Net, we can reduce the complexity of both the fixed and the moving images and iteratively derive their respective trajectories of gradually simplified images. The generated images at the ends of the two trajectories are so simple that they are very similar in appearance and morphology and thus easy to register. In this way, these two trajectories can act as a bridge to link the fixed and the moving images and guide their registration. Our experiments show that the proposed method can achieve more accurate registration results than state-of-the-art methods. Moreover, the proposed method can be generalized to the unseen dataset without the need for re-training or domain adaptation.",
keywords = "Brain MRI, Deep learning, Deformable registration, Morphology",
author = "Dongming Wei and Sahar Ahmad and Zhengwang Wu and Xiaohuan Cao and Xuhua Ren and Gang Li and Dinggang Shen and Qian Wang",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-32692-0_24",
language = "English",
isbn = "9783030326913",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
pages = "203--211",
editor = "Heung-Il Suk and Mingxia Liu and Chunfeng Lian and Pingkun Yan",
booktitle = "Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings",

}

TY - GEN

T1 - Morphological Simplification of Brain MR Images by Deep Learning for Facilitating Deformable Registration

AU - Wei, Dongming

AU - Ahmad, Sahar

AU - Wu, Zhengwang

AU - Cao, Xiaohuan

AU - Ren, Xuhua

AU - Li, Gang

AU - Shen, Dinggang

AU - Wang, Qian

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Brain MR image registration is challenging due to the large inter-subject anatomical variation. Especially, the highly convoluted brain cortex makes it difficult to accurately align the corresponding structures of the underlying images. In this paper, we propose a novel deep learning strategy to simplify the image registration task. Specifically, we train a morphological simplification network (MS-Net), which can generate a simplified image with fewer anatomical details given a complex input image. With this trained MS-Net, we can reduce the complexity of both the fixed and the moving images and iteratively derive their respective trajectories of gradually simplified images. The generated images at the ends of the two trajectories are so simple that they are very similar in appearance and morphology and thus easy to register. In this way, these two trajectories can act as a bridge to link the fixed and the moving images and guide their registration. Our experiments show that the proposed method can achieve more accurate registration results than state-of-the-art methods. Moreover, the proposed method can be generalized to the unseen dataset without the need for re-training or domain adaptation.

AB - Brain MR image registration is challenging due to the large inter-subject anatomical variation. Especially, the highly convoluted brain cortex makes it difficult to accurately align the corresponding structures of the underlying images. In this paper, we propose a novel deep learning strategy to simplify the image registration task. Specifically, we train a morphological simplification network (MS-Net), which can generate a simplified image with fewer anatomical details given a complex input image. With this trained MS-Net, we can reduce the complexity of both the fixed and the moving images and iteratively derive their respective trajectories of gradually simplified images. The generated images at the ends of the two trajectories are so simple that they are very similar in appearance and morphology and thus easy to register. In this way, these two trajectories can act as a bridge to link the fixed and the moving images and guide their registration. Our experiments show that the proposed method can achieve more accurate registration results than state-of-the-art methods. Moreover, the proposed method can be generalized to the unseen dataset without the need for re-training or domain adaptation.

KW - Brain MRI

KW - Deep learning

KW - Deformable registration

KW - Morphology

UR - http://www.scopus.com/inward/record.url?scp=85075673307&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075673307&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-32692-0_24

DO - 10.1007/978-3-030-32692-0_24

M3 - Conference contribution

AN - SCOPUS:85075673307

SN - 9783030326913

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

SP - 203

EP - 211

BT - Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings

A2 - Suk, Heung-Il

A2 - Liu, Mingxia

A2 - Lian, Chunfeng

A2 - Yan, Pingkun

PB - Springer

ER -