Multi-task Fully Convolutional Network for Brain Tumour Segmentation

Haocheng Shen; Ruixuan Wang; Jianguo Zhang; Stephen McKenna

doi:10.1007/978-3-319-60964-5_21

Multi-task Fully Convolutional Network for Brain Tumour Segmentation

Haocheng Shen (Lead / Corresponding author), Ruixuan Wang, Jianguo Zhang, Stephen McKenna

Computing

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

23 Citations (Scopus)

Abstract

In this paper, a novel, multi-task fully convolutional network (FCN) architecture is proposed for automatic segmentation of brain tumour. The proposed network builds on the hierarchical relationship between tumour substructures with branch and leaf losses imposed and optimised simultaneously. The network takes multimodal MR images along with their symmetric-difference images as input and extracts multi-level contextual information, firstly by the branch losses which are then fed to the leaf loss in a combination stage. The model was evaluated on BRATS13 and BRATS15 datasets and results show that the proposed multi-task FCN outperforms single-task FCN on all sub-tasks. The method is among the most accurate available and its computational cost is relatively low at test time.

Original language	English
Title of host publication	Medical Image Understanding and Analysis
Subtitle of host publication	21st Annual Conference, MIUA 2017, Edinburgh, UK, July 11–13, 2017, Proceedings
Editors	Maria Valdes Hernandez, Victor Gonzalez-Castro
Place of Publication	Switzerland
Publisher	Springer
Pages	239-248
Number of pages	10
ISBN (Electronic)	9783319609645
ISBN (Print)	9783319609638
DOIs	https://doi.org/10.1007/978-3-319-60964-5_21
Publication status	Published - 2017
Event	Medical Image Understanding and Analysis (MIUA) 2017 - John McIntyre Centre, Pollock Halls, Edinburgh, United Kingdom Duration: 11 Jul 2017 → 13 Jul 2017 https://miua2017.wordpress.com/

Publication series

Name	Communications in Computer and Information Science
Publisher	Springer
Volume	723
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	Medical Image Understanding and Analysis (MIUA) 2017
Abbreviated title	MIUA 2017
Country/Territory	United Kingdom
City	Edinburgh
Period	11/07/17 → 13/07/17
Internet address	https://miua2017.wordpress.com/

Keywords

Deep learning
Tumour segmentation
Multi-task learning

Access to Document

10.1007/978-3-319-60964-5_21

Cite this

Shen, H., Wang, R., Zhang, J., & McKenna, S. (2017). Multi-task Fully Convolutional Network for Brain Tumour Segmentation. In M. V. Hernandez, & V. Gonzalez-Castro (Eds.), Medical Image Understanding and Analysis: 21st Annual Conference, MIUA 2017, Edinburgh, UK, July 11–13, 2017, Proceedings (pp. 239-248). (Communications in Computer and Information Science; Vol. 723). Springer . https://doi.org/10.1007/978-3-319-60964-5_21

Shen, Haocheng ; Wang, Ruixuan ; Zhang, Jianguo et al. / Multi-task Fully Convolutional Network for Brain Tumour Segmentation. Medical Image Understanding and Analysis: 21st Annual Conference, MIUA 2017, Edinburgh, UK, July 11–13, 2017, Proceedings. editor / Maria Valdes Hernandez ; Victor Gonzalez-Castro. Switzerland : Springer , 2017. pp. 239-248 (Communications in Computer and Information Science).

@inproceedings{8776e261c3bf40638447a2ee62490bec,

title = "Multi-task Fully Convolutional Network for Brain Tumour Segmentation",

abstract = "In this paper, a novel, multi-task fully convolutional network (FCN) architecture is proposed for automatic segmentation of brain tumour. The proposed network builds on the hierarchical relationship between tumour substructures with branch and leaf losses imposed and optimised simultaneously. The network takes multimodal MR images along with their symmetric-difference images as input and extracts multi-level contextual information, firstly by the branch losses which are then fed to the leaf loss in a combination stage. The model was evaluated on BRATS13 and BRATS15 datasets and results show that the proposed multi-task FCN outperforms single-task FCN on all sub-tasks. The method is among the most accurate available and its computational cost is relatively low at test time.",

keywords = "Deep learning, Tumour segmentation, Multi-task learning",

author = "Haocheng Shen and Ruixuan Wang and Jianguo Zhang and Stephen McKenna",

note = "{\textcopyright} Springer International Publishing AG 2017; Medical Image Understanding and Analysis (MIUA) 2017, MIUA 2017 ; Conference date: 11-07-2017 Through 13-07-2017",

year = "2017",

doi = "10.1007/978-3-319-60964-5_21",

language = "English",

isbn = "9783319609638",

series = "Communications in Computer and Information Science",

publisher = "Springer ",

pages = "239--248",

editor = "Hernandez, {Maria Valdes} and Victor Gonzalez-Castro",

booktitle = "Medical Image Understanding and Analysis",

url = "https://miua2017.wordpress.com/",

}

Shen, H, Wang, R, Zhang, J & McKenna, S 2017, Multi-task Fully Convolutional Network for Brain Tumour Segmentation. in MV Hernandez & V Gonzalez-Castro (eds), Medical Image Understanding and Analysis: 21st Annual Conference, MIUA 2017, Edinburgh, UK, July 11–13, 2017, Proceedings. Communications in Computer and Information Science, vol. 723, Springer , Switzerland, pp. 239-248, Medical Image Understanding and Analysis (MIUA) 2017, Edinburgh, United Kingdom, 11/07/17. https://doi.org/10.1007/978-3-319-60964-5_21

Multi-task Fully Convolutional Network for Brain Tumour Segmentation. / Shen, Haocheng (Lead / Corresponding author); Wang, Ruixuan; Zhang, Jianguo et al.
Medical Image Understanding and Analysis: 21st Annual Conference, MIUA 2017, Edinburgh, UK, July 11–13, 2017, Proceedings. ed. / Maria Valdes Hernandez; Victor Gonzalez-Castro. Switzerland: Springer , 2017. p. 239-248 (Communications in Computer and Information Science; Vol. 723).

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

TY - GEN

T1 - Multi-task Fully Convolutional Network for Brain Tumour Segmentation

AU - Shen, Haocheng

AU - Wang, Ruixuan

AU - Zhang, Jianguo

AU - McKenna, Stephen

N1 - © Springer International Publishing AG 2017

PY - 2017

Y1 - 2017

N2 - In this paper, a novel, multi-task fully convolutional network (FCN) architecture is proposed for automatic segmentation of brain tumour. The proposed network builds on the hierarchical relationship between tumour substructures with branch and leaf losses imposed and optimised simultaneously. The network takes multimodal MR images along with their symmetric-difference images as input and extracts multi-level contextual information, firstly by the branch losses which are then fed to the leaf loss in a combination stage. The model was evaluated on BRATS13 and BRATS15 datasets and results show that the proposed multi-task FCN outperforms single-task FCN on all sub-tasks. The method is among the most accurate available and its computational cost is relatively low at test time.

AB - In this paper, a novel, multi-task fully convolutional network (FCN) architecture is proposed for automatic segmentation of brain tumour. The proposed network builds on the hierarchical relationship between tumour substructures with branch and leaf losses imposed and optimised simultaneously. The network takes multimodal MR images along with their symmetric-difference images as input and extracts multi-level contextual information, firstly by the branch losses which are then fed to the leaf loss in a combination stage. The model was evaluated on BRATS13 and BRATS15 datasets and results show that the proposed multi-task FCN outperforms single-task FCN on all sub-tasks. The method is among the most accurate available and its computational cost is relatively low at test time.

KW - Deep learning

KW - Tumour segmentation

KW - Multi-task learning

UR - https://miua2017.wordpress.com/programme/

U2 - 10.1007/978-3-319-60964-5_21

DO - 10.1007/978-3-319-60964-5_21

M3 - Conference contribution

SN - 9783319609638

T3 - Communications in Computer and Information Science

SP - 239

EP - 248

BT - Medical Image Understanding and Analysis

A2 - Hernandez, Maria Valdes

A2 - Gonzalez-Castro, Victor

PB - Springer

CY - Switzerland

T2 - Medical Image Understanding and Analysis (MIUA) 2017

Y2 - 11 July 2017 through 13 July 2017

ER -

Shen H, Wang R, Zhang J , McKenna S. Multi-task Fully Convolutional Network for Brain Tumour Segmentation. In Hernandez MV, Gonzalez-Castro V, editors, Medical Image Understanding and Analysis: 21st Annual Conference, MIUA 2017, Edinburgh, UK, July 11–13, 2017, Proceedings. Switzerland: Springer . 2017. p. 239-248. (Communications in Computer and Information Science). Epub 2017 Jun 22. doi: 10.1007/978-3-319-60964-5_21

Multi-task Fully Convolutional Network for Brain Tumour Segmentation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this