Deep learning based multimodal brain tumor diagnosis

Yuexiang Li; Linlin Shen

doi:10.1007/978-3-319-75238-9_13

Deep learning based multimodal brain tumor diagnosis

Yuexiang Li
, Linlin Shen

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

41 Citations (Scopus)

Abstract

Brain tumor segmentation plays an important role in the disease diagnosis. In this paper, we proposed deep learning frameworks, i.e. MvNet and SPNet, to address the challenges of multimodal brain tumor segmentation. The proposed multi-view deep learning framework (MvNet) uses three multi-branch fully-convolutional residual networks (Mb-FCRN) to segment multimodal brain images from different view-point, i.e. slices along x, y, z axis. The three sub-networks produce independent segmentation results and vote for the final outcome. The SPNet is a CNN-based framework developed to predict the survival time of patients. The proposed deep learning frameworks was evaluated on BraTS 17 validation set and achieved competing results for tumor segmentation While Dice scores of 0.88, 0.75 0.71 were achieved for whole tumor, enhancing tumor and tumor core, respectively, an accuracy of 0.55 was obtained for survival prediction.

Original language	English
Title of host publication	Brainlesion
Subtitle of host publication	Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 3rd International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017, Revised Selected Papers
Editors	Bjoern Menze, Alessandro Crimi, Hugo Kuijf, Mauricio Reyes, Spyridon Bakas
Publisher	Springer Verlag
Pages	149-158
Number of pages	10
ISBN (Print)	9783319752372
DOIs	https://doi.org/10.1007/978-3-319-75238-9_13
Publication status	Published - 2018
Externally published	Yes
Event	3rd International Workshop on Brainlesion, BrainLes 2017 Held in Conjunction with Medical Image Computing for Computer Assisted Intervention , MICCAI 2017 - Quebec City, Canada Duration: 14 Sept 2017 → 14 Sept 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10670 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	3rd International Workshop on Brainlesion, BrainLes 2017 Held in Conjunction with Medical Image Computing for Computer Assisted Intervention , MICCAI 2017
Country/Territory	Canada
City	Quebec City
Period	14/09/17 → 14/09/17

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Free Keywords

Deep learning
Multi-view
Survival prediction
Tumor segmentation

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-75238-9_13

Cite this

Li, Y., & Shen, L. (2018). Deep learning based multimodal brain tumor diagnosis. In B. Menze, A. Crimi, H. Kuijf, M. Reyes, & S. Bakas (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 3rd International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017, Revised Selected Papers (pp. 149-158). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10670 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-75238-9_13

Li, Yuexiang ; Shen, Linlin. / Deep learning based multimodal brain tumor diagnosis. Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 3rd International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017, Revised Selected Papers. editor / Bjoern Menze ; Alessandro Crimi ; Hugo Kuijf ; Mauricio Reyes ; Spyridon Bakas. Springer Verlag, 2018. pp. 149-158 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{1571433795f94f8aa4a11cf7e0192824,

title = "Deep learning based multimodal brain tumor diagnosis",

abstract = "Brain tumor segmentation plays an important role in the disease diagnosis. In this paper, we proposed deep learning frameworks, i.e. MvNet and SPNet, to address the challenges of multimodal brain tumor segmentation. The proposed multi-view deep learning framework (MvNet) uses three multi-branch fully-convolutional residual networks (Mb-FCRN) to segment multimodal brain images from different view-point, i.e. slices along x, y, z axis. The three sub-networks produce independent segmentation results and vote for the final outcome. The SPNet is a CNN-based framework developed to predict the survival time of patients. The proposed deep learning frameworks was evaluated on BraTS 17 validation set and achieved competing results for tumor segmentation While Dice scores of 0.88, 0.75 0.71 were achieved for whole tumor, enhancing tumor and tumor core, respectively, an accuracy of 0.55 was obtained for survival prediction.",

keywords = "Deep learning, Multi-view, Survival prediction, Tumor segmentation",

author = "Yuexiang Li and Linlin Shen",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG, part of Springer Nature 2018.; 3rd International Workshop on Brainlesion, BrainLes 2017 Held in Conjunction with Medical Image Computing for Computer Assisted Intervention , MICCAI 2017 ; Conference date: 14-09-2017 Through 14-09-2017",

year = "2018",

doi = "10.1007/978-3-319-75238-9\_13",

language = "English",

isbn = "9783319752372",

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

publisher = "Springer Verlag",

pages = "149--158",

editor = "Bjoern Menze and Alessandro Crimi and Hugo Kuijf and Mauricio Reyes and Spyridon Bakas",

booktitle = "Brainlesion",

address = "Germany",

}

Li, Y & Shen, L 2018, Deep learning based multimodal brain tumor diagnosis. in B Menze, A Crimi, H Kuijf, M Reyes & S Bakas (eds), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 3rd International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10670 LNCS, Springer Verlag, pp. 149-158, 3rd International Workshop on Brainlesion, BrainLes 2017 Held in Conjunction with Medical Image Computing for Computer Assisted Intervention , MICCAI 2017, Quebec City, Canada, 14/09/17. https://doi.org/10.1007/978-3-319-75238-9_13

Deep learning based multimodal brain tumor diagnosis. / Li, Yuexiang; Shen, Linlin.
Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 3rd International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017, Revised Selected Papers. ed. / Bjoern Menze; Alessandro Crimi; Hugo Kuijf; Mauricio Reyes; Spyridon Bakas. Springer Verlag, 2018. p. 149-158 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10670 LNCS).

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

TY - GEN

T1 - Deep learning based multimodal brain tumor diagnosis

AU - Li, Yuexiang

AU - Shen, Linlin

N1 - Publisher Copyright: © Springer International Publishing AG, part of Springer Nature 2018.

PY - 2018

Y1 - 2018

N2 - Brain tumor segmentation plays an important role in the disease diagnosis. In this paper, we proposed deep learning frameworks, i.e. MvNet and SPNet, to address the challenges of multimodal brain tumor segmentation. The proposed multi-view deep learning framework (MvNet) uses three multi-branch fully-convolutional residual networks (Mb-FCRN) to segment multimodal brain images from different view-point, i.e. slices along x, y, z axis. The three sub-networks produce independent segmentation results and vote for the final outcome. The SPNet is a CNN-based framework developed to predict the survival time of patients. The proposed deep learning frameworks was evaluated on BraTS 17 validation set and achieved competing results for tumor segmentation While Dice scores of 0.88, 0.75 0.71 were achieved for whole tumor, enhancing tumor and tumor core, respectively, an accuracy of 0.55 was obtained for survival prediction.

AB - Brain tumor segmentation plays an important role in the disease diagnosis. In this paper, we proposed deep learning frameworks, i.e. MvNet and SPNet, to address the challenges of multimodal brain tumor segmentation. The proposed multi-view deep learning framework (MvNet) uses three multi-branch fully-convolutional residual networks (Mb-FCRN) to segment multimodal brain images from different view-point, i.e. slices along x, y, z axis. The three sub-networks produce independent segmentation results and vote for the final outcome. The SPNet is a CNN-based framework developed to predict the survival time of patients. The proposed deep learning frameworks was evaluated on BraTS 17 validation set and achieved competing results for tumor segmentation While Dice scores of 0.88, 0.75 0.71 were achieved for whole tumor, enhancing tumor and tumor core, respectively, an accuracy of 0.55 was obtained for survival prediction.

KW - Deep learning

KW - Multi-view

KW - Survival prediction

KW - Tumor segmentation

UR - https://www.scopus.com/pages/publications/85042537633

U2 - 10.1007/978-3-319-75238-9_13

DO - 10.1007/978-3-319-75238-9_13

M3 - Conference contribution

AN - SCOPUS:85042537633

SN - 9783319752372

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

SP - 149

EP - 158

BT - Brainlesion

A2 - Menze, Bjoern

A2 - Crimi, Alessandro

A2 - Kuijf, Hugo

A2 - Reyes, Mauricio

A2 - Bakas, Spyridon

PB - Springer Verlag

T2 - 3rd International Workshop on Brainlesion, BrainLes 2017 Held in Conjunction with Medical Image Computing for Computer Assisted Intervention , MICCAI 2017

Y2 - 14 September 2017 through 14 September 2017

ER -

Li Y, Shen L. Deep learning based multimodal brain tumor diagnosis. In Menze B, Crimi A, Kuijf H, Reyes M, Bakas S, editors, Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 3rd International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017, Revised Selected Papers. Springer Verlag. 2018. p. 149-158. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-75238-9_13

Deep learning based multimodal brain tumor diagnosis

Abstract

Publication series

Conference

UN SDGs

Free Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this