Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Miaoyun Zhao, Li Wang, Jiawei Chen, Dong Nie, Yulai Cong, Sahar Ahmad, Angela Ho, Peng Yuan, Steve H. Fung, Hannah H. Deng, James J. Xia, Dinggang Shen

Automatic segmentation of medical images finds abundant applications in clinical studies. Computed Tomography (CT) imaging plays a critical role in diagnostic and surgical planning of craniomaxillofacial (CMF) surgeries as it shows clear bony structures. However, CT imaging poses radiation risks for the subjects being scanned. Alternatively, Magnetic Resonance Imaging (MRI) is considered to be safe and provides good visualization of the soft tissues, but the bony structures appear invisible from MRI. Therefore, the segmentation of bony structures from MRI is quite challenging. In this paper, we propose a cascaded generative adversarial network with deep-supervision discriminator (Deep-supGAN) for automatic bony structures segmentation. The first block in this architecture is used to generate a high-quality CT image from an MRI, and the second block is used to segment bony structures from MRI and the generated CT image. Different from traditional discriminators, the deep-supervision discriminator distinguishes the generated CT from the ground-truth at different levels of feature maps. For segmentation, the loss is not only concentrated on the voxel level but also on the higher abstract perceptual levels. Experimental results show that the proposed method generates CT images with clearer structural details and also segments the bony structures more accurately compared with the state-of-the-art methods.

Original languageEnglish (US)
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings
PublisherSpringer-Verlag
Pages720-727
Number of pages8
Volume11073
ISBN (Electronic)978-3-030-00937-3
ISBN (Print)978-3-030-00936-6
DOIs
StatePublished - Jan 1 2018
Event21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: Sep 16 2018Sep 20 2018

Publication series

NameMedical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

Other

Other21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period9/16/189/20/18

Altmetrics

Cite this

Standard

Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning. / Zhao, Miaoyun; Wang, Li; Chen, Jiawei; Nie, Dong; Cong, Yulai; Ahmad, Sahar; Ho, Angela; Yuan, Peng; Fung, Steve H.; Deng, Hannah H.; Xia, James J.; Shen, Dinggang.

Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. Vol. 11073 Springer-Verlag, 2018. p. 720-727 (Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Zhao, M, Wang, L, Chen, J, Nie, D, Cong, Y, Ahmad, S, Ho, A, Yuan, P, Fung, SH, Deng, HH, Xia, JJ & Shen, D 2018, Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning. in Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. vol. 11073, Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer-Verlag, pp. 720-727, 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018, Granada, Spain, 9/16/18. https://doi.org/10.1007/978-3-030-00937-3_82

APA

Zhao, M., Wang, L., Chen, J., Nie, D., Cong, Y., Ahmad, S., ... Shen, D. (2018). Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings (Vol. 11073, pp. 720-727). (Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention). Springer-Verlag. https://doi.org/10.1007/978-3-030-00937-3_82

Vancouver

Zhao M, Wang L, Chen J, Nie D, Cong Y, Ahmad S et al. Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. Vol. 11073. Springer-Verlag. 2018. p. 720-727. (Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention). https://doi.org/10.1007/978-3-030-00937-3_82

Author

Zhao, Miaoyun ; Wang, Li ; Chen, Jiawei ; Nie, Dong ; Cong, Yulai ; Ahmad, Sahar ; Ho, Angela ; Yuan, Peng ; Fung, Steve H. ; Deng, Hannah H. ; Xia, James J. ; Shen, Dinggang. / Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning. Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. Vol. 11073 Springer-Verlag, 2018. pp. 720-727 (Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention).

BibTeX

@inproceedings{bc5576207d21427bbc016cf1a278c326,
title = "Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning",
abstract = "Automatic segmentation of medical images finds abundant applications in clinical studies. Computed Tomography (CT) imaging plays a critical role in diagnostic and surgical planning of craniomaxillofacial (CMF) surgeries as it shows clear bony structures. However, CT imaging poses radiation risks for the subjects being scanned. Alternatively, Magnetic Resonance Imaging (MRI) is considered to be safe and provides good visualization of the soft tissues, but the bony structures appear invisible from MRI. Therefore, the segmentation of bony structures from MRI is quite challenging. In this paper, we propose a cascaded generative adversarial network with deep-supervision discriminator (Deep-supGAN) for automatic bony structures segmentation. The first block in this architecture is used to generate a high-quality CT image from an MRI, and the second block is used to segment bony structures from MRI and the generated CT image. Different from traditional discriminators, the deep-supervision discriminator distinguishes the generated CT from the ground-truth at different levels of feature maps. For segmentation, the loss is not only concentrated on the voxel level but also on the higher abstract perceptual levels. Experimental results show that the proposed method generates CT images with clearer structural details and also segments the bony structures more accurately compared with the state-of-the-art methods.",
author = "Miaoyun Zhao and Li Wang and Jiawei Chen and Dong Nie and Yulai Cong and Sahar Ahmad and Angela Ho and Peng Yuan and Fung, {Steve H.} and Deng, {Hannah H.} and Xia, {James J.} and Dinggang Shen",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-030-00937-3_82",
language = "English (US)",
isbn = "978-3-030-00936-6",
volume = "11073",
series = "Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention",
publisher = "Springer-Verlag",
pages = "720--727",
booktitle = "Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings",

}

RIS

TY - GEN

T1 - Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning

AU - Zhao, Miaoyun

AU - Wang, Li

AU - Chen, Jiawei

AU - Nie, Dong

AU - Cong, Yulai

AU - Ahmad, Sahar

AU - Ho, Angela

AU - Yuan, Peng

AU - Fung, Steve H.

AU - Deng, Hannah H.

AU - Xia, James J.

AU - Shen, Dinggang

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Automatic segmentation of medical images finds abundant applications in clinical studies. Computed Tomography (CT) imaging plays a critical role in diagnostic and surgical planning of craniomaxillofacial (CMF) surgeries as it shows clear bony structures. However, CT imaging poses radiation risks for the subjects being scanned. Alternatively, Magnetic Resonance Imaging (MRI) is considered to be safe and provides good visualization of the soft tissues, but the bony structures appear invisible from MRI. Therefore, the segmentation of bony structures from MRI is quite challenging. In this paper, we propose a cascaded generative adversarial network with deep-supervision discriminator (Deep-supGAN) for automatic bony structures segmentation. The first block in this architecture is used to generate a high-quality CT image from an MRI, and the second block is used to segment bony structures from MRI and the generated CT image. Different from traditional discriminators, the deep-supervision discriminator distinguishes the generated CT from the ground-truth at different levels of feature maps. For segmentation, the loss is not only concentrated on the voxel level but also on the higher abstract perceptual levels. Experimental results show that the proposed method generates CT images with clearer structural details and also segments the bony structures more accurately compared with the state-of-the-art methods.

AB - Automatic segmentation of medical images finds abundant applications in clinical studies. Computed Tomography (CT) imaging plays a critical role in diagnostic and surgical planning of craniomaxillofacial (CMF) surgeries as it shows clear bony structures. However, CT imaging poses radiation risks for the subjects being scanned. Alternatively, Magnetic Resonance Imaging (MRI) is considered to be safe and provides good visualization of the soft tissues, but the bony structures appear invisible from MRI. Therefore, the segmentation of bony structures from MRI is quite challenging. In this paper, we propose a cascaded generative adversarial network with deep-supervision discriminator (Deep-supGAN) for automatic bony structures segmentation. The first block in this architecture is used to generate a high-quality CT image from an MRI, and the second block is used to segment bony structures from MRI and the generated CT image. Different from traditional discriminators, the deep-supervision discriminator distinguishes the generated CT from the ground-truth at different levels of feature maps. For segmentation, the loss is not only concentrated on the voxel level but also on the higher abstract perceptual levels. Experimental results show that the proposed method generates CT images with clearer structural details and also segments the bony structures more accurately compared with the state-of-the-art methods.

UR - http://www.scopus.com/inward/record.url?scp=85053840010&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053840010&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00937-3_82

DO - 10.1007/978-3-030-00937-3_82

M3 - Conference contribution

SN - 978-3-030-00936-6

VL - 11073

T3 - Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

SP - 720

EP - 727

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings

PB - Springer-Verlag

ER -

ID: 40990900