Multi-scale networks for segmentation of brain magnetic resonance images

Jie Wei; Yong Xia

doi:10.1007/978-3-319-67558-9_36

Multi-scale networks for segmentation of brain magnetic resonance images

Jie Wei, Yong Xia

School of Computer Science

Northwestern Polytechnical University Xian

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Measuring the distribution of major brain tissues using magnetic resonance (MR) images has attracted extensive research efforts. Due to its remarkable success, deep learning-based image segmentation has been applied to this problem, in which the size of patches usually represents a tradeoff between complexity and accuracy. In this paper, we propose the multi-size-and-position neural network (MSPNN) for brain MR image segmentation. Our contributions include (1) jointly using U-Nets trained on large patches and back propagation neural networks (BPNNs) trained on small patches for segmentation, and (2) adopting the convolutional auto-encoder (CAE) to restore MR images before applying them to BPNNs. We have evaluated this algorithm against five widely used brain MR image segmentation approaches on both synthetic and real MR studies. Our results indicate that the proposed algorithm can segment brain MR images effectively and provide precise distribution of major brain tissues.

Original language	English
Title of host publication	Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings
Editors	Tal Arbel, M. Jorge Cardoso
Publisher	Springer Verlag
Pages	312-320
Number of pages	9
ISBN (Print)	9783319675572
DOIs	https://doi.org/10.1007/978-3-319-67558-9_36
State	Published - 2017
Event	3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 - Quebec City, Canada Duration: 14 Sep 2017 → 14 Sep 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10553 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017
Country/Territory	Canada
City	Quebec City
Period	14/09/17 → 14/09/17

Keywords

Back propagation neural network (BPNN)
Brain MR image segmentation
Convolutional auto-encoder (CAE)
Deep learning
U-Net

Access to Document

10.1007/978-3-319-67558-9_36

Cite this

Wei, J., & Xia, Y. (2017). Multi-scale networks for segmentation of brain magnetic resonance images. In T. Arbel, & M. J. Cardoso (Eds.), Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings (pp. 312-320). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10553 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-67558-9_36

Wei, Jie ; Xia, Yong. / Multi-scale networks for segmentation of brain magnetic resonance images. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. editor / Tal Arbel ; M. Jorge Cardoso. Springer Verlag, 2017. pp. 312-320 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{86a6d3050b964bf2a80f840d70f3614a,

title = "Multi-scale networks for segmentation of brain magnetic resonance images",

abstract = "Measuring the distribution of major brain tissues using magnetic resonance (MR) images has attracted extensive research efforts. Due to its remarkable success, deep learning-based image segmentation has been applied to this problem, in which the size of patches usually represents a tradeoff between complexity and accuracy. In this paper, we propose the multi-size-and-position neural network (MSPNN) for brain MR image segmentation. Our contributions include (1) jointly using U-Nets trained on large patches and back propagation neural networks (BPNNs) trained on small patches for segmentation, and (2) adopting the convolutional auto-encoder (CAE) to restore MR images before applying them to BPNNs. We have evaluated this algorithm against five widely used brain MR image segmentation approaches on both synthetic and real MR studies. Our results indicate that the proposed algorithm can segment brain MR images effectively and provide precise distribution of major brain tissues.",

keywords = "Back propagation neural network (BPNN), Brain MR image segmentation, Convolutional auto-encoder (CAE), Deep learning, U-Net",

author = "Jie Wei and Yong Xia",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 ; Conference date: 14-09-2017 Through 14-09-2017",

year = "2017",

doi = "10.1007/978-3-319-67558-9_36",

language = "英语",

isbn = "9783319675572",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "312--320",

editor = "Tal Arbel and Cardoso, {M. Jorge}",

booktitle = "Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings",

}

Wei, J & Xia, Y 2017, Multi-scale networks for segmentation of brain magnetic resonance images. in T Arbel & MJ Cardoso (eds), Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10553 LNCS, Springer Verlag, pp. 312-320, 3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, Quebec City, Canada, 14/09/17. https://doi.org/10.1007/978-3-319-67558-9_36

Multi-scale networks for segmentation of brain magnetic resonance images. / Wei, Jie; Xia, Yong.
Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. ed. / Tal Arbel; M. Jorge Cardoso. Springer Verlag, 2017. p. 312-320 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10553 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Multi-scale networks for segmentation of brain magnetic resonance images

AU - Wei, Jie

AU - Xia, Yong

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - Measuring the distribution of major brain tissues using magnetic resonance (MR) images has attracted extensive research efforts. Due to its remarkable success, deep learning-based image segmentation has been applied to this problem, in which the size of patches usually represents a tradeoff between complexity and accuracy. In this paper, we propose the multi-size-and-position neural network (MSPNN) for brain MR image segmentation. Our contributions include (1) jointly using U-Nets trained on large patches and back propagation neural networks (BPNNs) trained on small patches for segmentation, and (2) adopting the convolutional auto-encoder (CAE) to restore MR images before applying them to BPNNs. We have evaluated this algorithm against five widely used brain MR image segmentation approaches on both synthetic and real MR studies. Our results indicate that the proposed algorithm can segment brain MR images effectively and provide precise distribution of major brain tissues.

AB - Measuring the distribution of major brain tissues using magnetic resonance (MR) images has attracted extensive research efforts. Due to its remarkable success, deep learning-based image segmentation has been applied to this problem, in which the size of patches usually represents a tradeoff between complexity and accuracy. In this paper, we propose the multi-size-and-position neural network (MSPNN) for brain MR image segmentation. Our contributions include (1) jointly using U-Nets trained on large patches and back propagation neural networks (BPNNs) trained on small patches for segmentation, and (2) adopting the convolutional auto-encoder (CAE) to restore MR images before applying them to BPNNs. We have evaluated this algorithm against five widely used brain MR image segmentation approaches on both synthetic and real MR studies. Our results indicate that the proposed algorithm can segment brain MR images effectively and provide precise distribution of major brain tissues.

KW - Back propagation neural network (BPNN)

KW - Brain MR image segmentation

KW - Convolutional auto-encoder (CAE)

KW - Deep learning

KW - U-Net

UR - http://www.scopus.com/inward/record.url?scp=85029816004&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-67558-9_36

DO - 10.1007/978-3-319-67558-9_36

M3 - 会议稿件

AN - SCOPUS:85029816004

SN - 9783319675572

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

SP - 312

EP - 320

BT - Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings

A2 - Arbel, Tal

A2 - Cardoso, M. Jorge

PB - Springer Verlag

T2 - 3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017

Y2 - 14 September 2017 through 14 September 2017

ER -

Wei J, Xia Y. Multi-scale networks for segmentation of brain magnetic resonance images. In Arbel T, Cardoso MJ, editors, Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. Springer Verlag. 2017. p. 312-320. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-67558-9_36

Multi-scale networks for segmentation of brain magnetic resonance images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this