Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation

Jingkun Chen; Hongwei Li; Jianguo Zhang; Bjoern Menze

doi:10.1007/978-3-030-39074-7_34

Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation

Jingkun Chen, Hongwei Li, Jianguo Zhang, Bjoern Menze

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

13 Citations (Scopus)

166 Downloads (Pure)

Abstract

Analysis and modeling of the ventricles and myocardium are important in the diagnostic and treatment of heart diseases. Manual delineation of those tissues in cardiac MR (CMR) scans is laborious and time-consuming. The ambiguity of the boundaries makes the segmentation task rather challenging. Furthermore, the annotations on some modalities such as Late Gadolinium Enhancement (LGE) MRI, are often not available. We propose an end-to-end segmentation framework based on convolutional neural network (CNN) and adversarial learning. A dilated residual U-shape network is used as a segmentor to generate the prediction mask; meanwhile, a CNN is utilized as a discriminator model to judge the segmentation quality. To leverage the available annotations across modalities per patient, a new loss function named weak domain-transfer loss is introduced to the pipeline. The proposed model is evaluated on the public dataset released by the challenge organizer in MICCAI 2019, which consists of 45 sets of multi-sequence CMR images. We demonstrate that the proposed adversarial pipeline outperforms baseline deep-learning methods.

Original language	English
Title of host publication	Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges
Subtitle of host publication	10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers
Editors	Mihaela Pop, Maxime Sermesant, Oscar Camara, Xiahai Zhuang, Shuo Li, Alistair Young, Tommaso Mansi, Avan Suinesiaputra
Publisher	Springer Verlag
Pages	317-325
Number of pages	9
Volume	12009
ISBN (Electronic)	9783030390747
ISBN (Print)	9783030390730
DOIs	https://doi.org/10.1007/978-3-030-39074-7_34
Publication status	Published - 2020
Event	10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019 - Shenzhen, China Duration: 13 Oct 2019 → 13 Oct 2019

Publication series

Name	Lecture Notes in Computer Science
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019
Country/Territory	China
City	Shenzhen
Period	13/10/19 → 13/10/19

Keywords

Adversarial convolutional network
Multi-sequence cardiac segmentation

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-030-39074-7_34

Author Accepted ManuscriptAccepted author manuscript, 1.62 MB

Cite this

Chen, J., Li, H., Zhang, J., & Menze, B. (2020). Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation. In M. Pop, M. Sermesant, O. Camara, X. Zhuang, S. Li, A. Young, T. Mansi, & A. Suinesiaputra (Eds.), Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges: 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers (Vol. 12009, pp. 317-325). (Lecture Notes in Computer Science). Springer Verlag. https://doi.org/10.1007/978-3-030-39074-7_34

Chen, Jingkun ; Li, Hongwei ; Zhang, Jianguo et al. / Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation. Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges: 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. editor / Mihaela Pop ; Maxime Sermesant ; Oscar Camara ; Xiahai Zhuang ; Shuo Li ; Alistair Young ; Tommaso Mansi ; Avan Suinesiaputra. Vol. 12009 Springer Verlag, 2020. pp. 317-325 (Lecture Notes in Computer Science).

@inproceedings{a013a611d76d4a2d87a8697a2b6ad991,

title = "Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation",

abstract = "Analysis and modeling of the ventricles and myocardium are important in the diagnostic and treatment of heart diseases. Manual delineation of those tissues in cardiac MR (CMR) scans is laborious and time-consuming. The ambiguity of the boundaries makes the segmentation task rather challenging. Furthermore, the annotations on some modalities such as Late Gadolinium Enhancement (LGE) MRI, are often not available. We propose an end-to-end segmentation framework based on convolutional neural network (CNN) and adversarial learning. A dilated residual U-shape network is used as a segmentor to generate the prediction mask; meanwhile, a CNN is utilized as a discriminator model to judge the segmentation quality. To leverage the available annotations across modalities per patient, a new loss function named weak domain-transfer loss is introduced to the pipeline. The proposed model is evaluated on the public dataset released by the challenge organizer in MICCAI 2019, which consists of 45 sets of multi-sequence CMR images. We demonstrate that the proposed adversarial pipeline outperforms baseline deep-learning methods.",

keywords = "Adversarial convolutional network, Multi-sequence cardiac segmentation",

author = "Jingkun Chen and Hongwei Li and Jianguo Zhang and Bjoern Menze",

year = "2020",

doi = "10.1007/978-3-030-39074-7_34",

language = "English",

isbn = "9783030390730",

volume = "12009",

series = "Lecture Notes in Computer Science",

publisher = "Springer Verlag",

pages = "317--325",

editor = "Mihaela Pop and Maxime Sermesant and Oscar Camara and Xiahai Zhuang and Shuo Li and Alistair Young and Tommaso Mansi and Avan Suinesiaputra",

booktitle = "Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges",

address = "Germany",

note = "10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019 ; Conference date: 13-10-2019 Through 13-10-2019",

}

Chen, J, Li, H , Zhang, J & Menze, B 2020, Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation. in M Pop, M Sermesant, O Camara, X Zhuang, S Li, A Young, T Mansi & A Suinesiaputra (eds), Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges: 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. vol. 12009, Lecture Notes in Computer Science, Springer Verlag, pp. 317-325, 10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019, Shenzhen, China, 13/10/19. https://doi.org/10.1007/978-3-030-39074-7_34

Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation. / Chen, Jingkun; Li, Hongwei ; Zhang, Jianguo et al.
Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges: 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. ed. / Mihaela Pop; Maxime Sermesant; Oscar Camara; Xiahai Zhuang; Shuo Li; Alistair Young; Tommaso Mansi; Avan Suinesiaputra. Vol. 12009 Springer Verlag, 2020. p. 317-325 (Lecture Notes in Computer Science).

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

TY - GEN

T1 - Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation

AU - Chen, Jingkun

AU - Li, Hongwei

AU - Zhang, Jianguo

AU - Menze, Bjoern

PY - 2020

Y1 - 2020

N2 - Analysis and modeling of the ventricles and myocardium are important in the diagnostic and treatment of heart diseases. Manual delineation of those tissues in cardiac MR (CMR) scans is laborious and time-consuming. The ambiguity of the boundaries makes the segmentation task rather challenging. Furthermore, the annotations on some modalities such as Late Gadolinium Enhancement (LGE) MRI, are often not available. We propose an end-to-end segmentation framework based on convolutional neural network (CNN) and adversarial learning. A dilated residual U-shape network is used as a segmentor to generate the prediction mask; meanwhile, a CNN is utilized as a discriminator model to judge the segmentation quality. To leverage the available annotations across modalities per patient, a new loss function named weak domain-transfer loss is introduced to the pipeline. The proposed model is evaluated on the public dataset released by the challenge organizer in MICCAI 2019, which consists of 45 sets of multi-sequence CMR images. We demonstrate that the proposed adversarial pipeline outperforms baseline deep-learning methods.

AB - Analysis and modeling of the ventricles and myocardium are important in the diagnostic and treatment of heart diseases. Manual delineation of those tissues in cardiac MR (CMR) scans is laborious and time-consuming. The ambiguity of the boundaries makes the segmentation task rather challenging. Furthermore, the annotations on some modalities such as Late Gadolinium Enhancement (LGE) MRI, are often not available. We propose an end-to-end segmentation framework based on convolutional neural network (CNN) and adversarial learning. A dilated residual U-shape network is used as a segmentor to generate the prediction mask; meanwhile, a CNN is utilized as a discriminator model to judge the segmentation quality. To leverage the available annotations across modalities per patient, a new loss function named weak domain-transfer loss is introduced to the pipeline. The proposed model is evaluated on the public dataset released by the challenge organizer in MICCAI 2019, which consists of 45 sets of multi-sequence CMR images. We demonstrate that the proposed adversarial pipeline outperforms baseline deep-learning methods.

KW - Adversarial convolutional network

KW - Multi-sequence cardiac segmentation

UR - http://www.scopus.com/inward/record.url?scp=85081932812&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-39074-7_34

DO - 10.1007/978-3-030-39074-7_34

M3 - Conference contribution

AN - SCOPUS:85081932812

SN - 9783030390730

VL - 12009

T3 - Lecture Notes in Computer Science

SP - 317

EP - 325

BT - Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges

A2 - Pop, Mihaela

A2 - Sermesant, Maxime

A2 - Camara, Oscar

A2 - Zhuang, Xiahai

A2 - Li, Shuo

A2 - Young, Alistair

A2 - Mansi, Tommaso

A2 - Suinesiaputra, Avan

PB - Springer Verlag

T2 - 10th International Workshop on Statistical Atlases and Computational Models of the Heart, STACOM 2019, held in conjunction with the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2019

Y2 - 13 October 2019 through 13 October 2019

ER -

Chen J, Li H , Zhang J, Menze B. Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-sequence Cardiac MR Images Segmentation. In Pop M, Sermesant M, Camara O, Zhuang X, Li S, Young A, Mansi T, Suinesiaputra A, editors, Statistical Atlases and Computational Models of the Heart. Multi-Sequence CMR Segmentation, CRT-EPiggy and LV Full Quantification Challenges: 10th International Workshop, STACOM 2019, Held in Conjunction with MICCAI 2019, Revised Selected Papers. Vol. 12009. Springer Verlag. 2020. p. 317-325. (Lecture Notes in Computer Science). Epub 2020 Jan 23. doi: 10.1007/978-3-030-39074-7_34