Adversarial training and dilated convolutions for brain MRI segmentation

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

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

40 Citaten (Scopus)
2 Downloads (Pure)

Samenvatting

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.

Originele taal-2Engels
TitelDeep 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
UitgeverijSpringer
Pagina's56-64
Aantal pagina's9
ISBN van geprinte versie9783319675572
DOI's
StatusGepubliceerd - 2017
Evenement3rd 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
Duur: 10 sep 201714 sep 2017
http://www.miccai2017.org/

Publicatie series

NaamLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10553 LNCS
ISSN van geprinte versie0302-9743
ISSN van elektronische versie1611-3349

Congres

Congres3rd 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
Verkorte titelDLMIA2017
LandCanada
StadQuebec City
Periode10/09/1714/09/17
Internet adres

Vingerafdruk Duik in de onderzoeksthema's van 'Adversarial training and dilated convolutions for brain MRI segmentation'. Samen vormen ze een unieke vingerafdruk.

  • Citeer dit

    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 (blz. 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