Adversarial training and dilated convolutions for brain MRI segmentation

P. Moeskops, M. Veta, M.W. Lafarge, K.A.J. Eppenhof, J.P.W. Pluim

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

35 Citations (Scopus)
2 Downloads (Pure)

Abstract

Convolutional neural networks (CNNs) have been applied to various automatic image segmentation tasks in medical image analysis, including brain MRI segmentation. Generative adversarial networks have recently gained popularity because of their power in generating images that are difficult to distinguish from real images. In this study we use an adversarial training approach to improve CNN-based brain MRI segmentation. To this end, we include an additional loss function that motivates the network to generate segmentations that are difficult to distinguish from manual segmentations. During training, this loss function is optimised together with the conventional average per-voxel cross entropy loss. The results show improved segmentation performance using this adversarial training procedure for segmentation of two different sets of images and using two different network architectures, both visually and in terms of Dice coefficients.

Original languageEnglish
Title of host publicationDeep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings
PublisherSpringer
Pages56-64
Number of pages9
ISBN (Print)9783319675572
DOIs
Publication statusPublished - 2017
Event3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, 10-14 September Quebec, Canada - Quebec City, Canada
Duration: 10 Sep 201714 Sep 2017
http://www.miccai2017.org/

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10553 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, 10-14 September Quebec, Canada
Abbreviated titleDLMIA2017
CountryCanada
CityQuebec City
Period10/09/1714/09/17
Internet address

Keywords

  • Adversarial networks
  • Brain MRI
  • Convolutional neural networks
  • Deep learning
  • Dilated convolution
  • Medical image segmentation

Fingerprint Dive into the research topics of 'Adversarial training and dilated convolutions for brain MRI segmentation'. Together they form a unique fingerprint.

  • Cite this

    Moeskops, P., Veta, M., Lafarge, M. W., Eppenhof, K. A. J., & Pluim, J. P. W. (2017). Adversarial training and dilated convolutions for brain MRI segmentation. In Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings (pp. 56-64). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10553 LNCS). Springer. https://doi.org/10.1007/978-3-319-67558-9_7