3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge

Yoonseok Choi; Mohammed A. Al-masni; Dong Hyun Kim

doi:10.1007/978-3-031-08999-2_28

3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge

Yoonseok Choi, Mohammed A. Al-masni, Dong Hyun Kim

Department of Artificial Intelligence

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

Abstract

We propose a 3D version of the Contextual Multi-scale Multi-level Network (3D CMM-Net) with deeper encoder depth for automated semantic segmentation of different brain tumors in the BraTS2021 challenge. The proposed network has the capability to extract and learn deeper features for the task of multi-class segmentation directly from 3D MRI data. The overall performance of the proposed network gave Dice scores of 0.7557, 0.8060, and 0.8351 for enhancing tumor, tumor core, and whole tumor, respectively on the local-test dataset.

Original language	English
Title of host publication	Brainlesion
Subtitle of host publication	Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 7th International Workshop, BrainLes 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers
Editors	Alessandro Crimi, Spyridon Bakas
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	333-343
Number of pages	11
ISBN (Print)	9783031089985
DOIs	https://doi.org/10.1007/978-3-031-08999-2_28
State	Published - 2022
Event	7th International Brain Lesion Workshop, BrainLes 2021, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 - Virtual, Online Duration: 27 Sep 2021 → 27 Sep 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12962 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	7th International Brain Lesion Workshop, BrainLes 2021, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2021
City	Virtual, Online
Period	27/09/21 → 27/09/21

Keywords

3D semantic segmentation
Brain tumor segmentation
Glioblastoma
Multimodal MRI
Pyramid pooling module
U-Net

Access to Document

10.1007/978-3-031-08999-2_28

Cite this

Choi, Y., Al-masni, M. A., & Kim, D. H. (2022). 3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge. In A. Crimi, & S. Bakas (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 7th International Workshop, BrainLes 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers (pp. 333-343). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12962 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-08999-2_28

Choi, Yoonseok ; Al-masni, Mohammed A. ; Kim, Dong Hyun. / 3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge. Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 7th International Workshop, BrainLes 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. editor / Alessandro Crimi ; Spyridon Bakas. Springer Science and Business Media Deutschland GmbH, 2022. pp. 333-343 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{6c8417804376408ca6c7e50dab5f5765,

title = "3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge",

abstract = "We propose a 3D version of the Contextual Multi-scale Multi-level Network (3D CMM-Net) with deeper encoder depth for automated semantic segmentation of different brain tumors in the BraTS2021 challenge. The proposed network has the capability to extract and learn deeper features for the task of multi-class segmentation directly from 3D MRI data. The overall performance of the proposed network gave Dice scores of 0.7557, 0.8060, and 0.8351 for enhancing tumor, tumor core, and whole tumor, respectively on the local-test dataset.",

keywords = "3D semantic segmentation, Brain tumor segmentation, Glioblastoma, Multimodal MRI, Pyramid pooling module, U-Net",

author = "Yoonseok Choi and Al-masni, {Mohammed A.} and Kim, {Dong Hyun}",

note = "Funding Information: Acknowledgements. This work was supported by the Korea Medical Device Development Fund grant funded by the Korea government (the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, Republic of Korea, the Ministry of Food and Drug Safety) (Project Number: 202011D23). Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; null ; Conference date: 27-09-2021 Through 27-09-2021",

year = "2022",

doi = "10.1007/978-3-031-08999-2_28",

language = "English",

isbn = "9783031089985",

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

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "333--343",

editor = "Alessandro Crimi and Spyridon Bakas",

booktitle = "Brainlesion",

}

Choi, Y, Al-masni, MA & Kim, DH 2022, 3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge. in A Crimi & S Bakas (eds), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 7th International Workshop, BrainLes 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12962 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 333-343, 7th International Brain Lesion Workshop, BrainLes 2021, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2021, Virtual, Online, 27/09/21. https://doi.org/10.1007/978-3-031-08999-2_28

3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge. / Choi, Yoonseok; Al-masni, Mohammed A.; Kim, Dong Hyun.

Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 7th International Workshop, BrainLes 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. ed. / Alessandro Crimi; Spyridon Bakas. Springer Science and Business Media Deutschland GmbH, 2022. p. 333-343 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12962 LNCS).

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

TY - GEN

T1 - 3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge

AU - Choi, Yoonseok

AU - Al-masni, Mohammed A.

AU - Kim, Dong Hyun

N1 - Funding Information: Acknowledgements. This work was supported by the Korea Medical Device Development Fund grant funded by the Korea government (the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, Republic of Korea, the Ministry of Food and Drug Safety) (Project Number: 202011D23). Publisher Copyright: © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

PY - 2022

Y1 - 2022

N2 - We propose a 3D version of the Contextual Multi-scale Multi-level Network (3D CMM-Net) with deeper encoder depth for automated semantic segmentation of different brain tumors in the BraTS2021 challenge. The proposed network has the capability to extract and learn deeper features for the task of multi-class segmentation directly from 3D MRI data. The overall performance of the proposed network gave Dice scores of 0.7557, 0.8060, and 0.8351 for enhancing tumor, tumor core, and whole tumor, respectively on the local-test dataset.

AB - We propose a 3D version of the Contextual Multi-scale Multi-level Network (3D CMM-Net) with deeper encoder depth for automated semantic segmentation of different brain tumors in the BraTS2021 challenge. The proposed network has the capability to extract and learn deeper features for the task of multi-class segmentation directly from 3D MRI data. The overall performance of the proposed network gave Dice scores of 0.7557, 0.8060, and 0.8351 for enhancing tumor, tumor core, and whole tumor, respectively on the local-test dataset.

KW - 3D semantic segmentation

KW - Brain tumor segmentation

KW - Glioblastoma

KW - Multimodal MRI

KW - Pyramid pooling module

KW - U-Net

UR - http://www.scopus.com/inward/record.url?scp=85135050566&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-08999-2_28

DO - 10.1007/978-3-031-08999-2_28

M3 - Conference contribution

AN - SCOPUS:85135050566

SN - 9783031089985

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

SP - 333

EP - 343

BT - Brainlesion

A2 - Crimi, Alessandro

A2 - Bakas, Spyridon

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 27 September 2021 through 27 September 2021

ER -

Choi Y, Al-masni MA, Kim DH. 3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge. In Crimi A, Bakas S, editors, Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 7th International Workshop, BrainLes 2021, Held in Conjunction with MICCAI 2021, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2022. p. 333-343. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-08999-2_28

3D CMM-Net with Deeper Encoder for Semantic Segmentation of Brain Tumors in BraTS2021 Challenge

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this