Brain image labeling using multi-atlas guided 3D fully convolutional networks

Longwei Fang, Lichi Zhang, Dong Nie, Xiaohuan Cao, Khosro Bahrami, Huiguang He, Dinggang Shen

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

4 Citations (Scopus)

Abstract

Automatic labeling of anatomical structures in brain images plays an important role in neuroimaging analysis. Among all methods, multi-atlas based segmentation methods are widely used, due to their robustness in propagating prior label information. However, non-linear registration is always needed, which is time-consuming. Alternatively, the patch-based methods have been proposed to relax the requirement of image registration, but the labeling is often determined independently by the target image information, without getting direct assistance from the atlases. To address these limitations, in this paper, we propose a multi-atlas guided 3D fully convolutional networks (FCN) for brain image labeling. Specifically, multi-atlas based guidance is incorporated during the network learning. Based on this, the discriminative of the FCN is boosted, which eventually contribute to accurate prediction. Experiments show that the use of multi-atlas guidance improves the brain labeling performance.

Original languageEnglish
Title of host publicationPatch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings
PublisherSpringer Verlag
Pages12-19
Number of pages8
Volume10530 LNCS
ISBN (Print)9783319674339
DOIs
Publication statusPublished - 2017
Externally publishedYes
Event3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 - Quebec City, Canada
Duration: 2017 Sep 142017 Sep 14

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10530 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017
CountryCanada
CityQuebec City
Period17/9/1417/9/14

Fingerprint

Atlas
Labeling
Brain
Guidance
Neuroimaging
Image registration
Labels
Image Registration
Registration
Patch
Segmentation
Robustness
Target
Prediction
Requirements
Experiments
Experiment

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Fang, L., Zhang, L., Nie, D., Cao, X., Bahrami, K., He, H., & Shen, D. (2017). Brain image labeling using multi-atlas guided 3D fully convolutional networks. In Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings (Vol. 10530 LNCS, pp. 12-19). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10530 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-67434-6_2

Brain image labeling using multi-atlas guided 3D fully convolutional networks. / Fang, Longwei; Zhang, Lichi; Nie, Dong; Cao, Xiaohuan; Bahrami, Khosro; He, Huiguang; Shen, Dinggang.

Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10530 LNCS Springer Verlag, 2017. p. 12-19 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10530 LNCS).

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

Fang, L, Zhang, L, Nie, D, Cao, X, Bahrami, K, He, H & Shen, D 2017, Brain image labeling using multi-atlas guided 3D fully convolutional networks. in Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. vol. 10530 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10530 LNCS, Springer Verlag, pp. 12-19, 3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, Quebec City, Canada, 17/9/14. https://doi.org/10.1007/978-3-319-67434-6_2
Fang L, Zhang L, Nie D, Cao X, Bahrami K, He H et al. Brain image labeling using multi-atlas guided 3D fully convolutional networks. In Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10530 LNCS. Springer Verlag. 2017. p. 12-19. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-67434-6_2
Fang, Longwei ; Zhang, Lichi ; Nie, Dong ; Cao, Xiaohuan ; Bahrami, Khosro ; He, Huiguang ; Shen, Dinggang. / Brain image labeling using multi-atlas guided 3D fully convolutional networks. Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10530 LNCS Springer Verlag, 2017. pp. 12-19 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{1a060d1c88f34ba1be07618d9434885e,
title = "Brain image labeling using multi-atlas guided 3D fully convolutional networks",
abstract = "Automatic labeling of anatomical structures in brain images plays an important role in neuroimaging analysis. Among all methods, multi-atlas based segmentation methods are widely used, due to their robustness in propagating prior label information. However, non-linear registration is always needed, which is time-consuming. Alternatively, the patch-based methods have been proposed to relax the requirement of image registration, but the labeling is often determined independently by the target image information, without getting direct assistance from the atlases. To address these limitations, in this paper, we propose a multi-atlas guided 3D fully convolutional networks (FCN) for brain image labeling. Specifically, multi-atlas based guidance is incorporated during the network learning. Based on this, the discriminative of the FCN is boosted, which eventually contribute to accurate prediction. Experiments show that the use of multi-atlas guidance improves the brain labeling performance.",
author = "Longwei Fang and Lichi Zhang and Dong Nie and Xiaohuan Cao and Khosro Bahrami and Huiguang He and Dinggang Shen",
year = "2017",
doi = "10.1007/978-3-319-67434-6_2",
language = "English",
isbn = "9783319674339",
volume = "10530 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "12--19",
booktitle = "Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings",

}

TY - GEN

T1 - Brain image labeling using multi-atlas guided 3D fully convolutional networks

AU - Fang, Longwei

AU - Zhang, Lichi

AU - Nie, Dong

AU - Cao, Xiaohuan

AU - Bahrami, Khosro

AU - He, Huiguang

AU - Shen, Dinggang

PY - 2017

Y1 - 2017

N2 - Automatic labeling of anatomical structures in brain images plays an important role in neuroimaging analysis. Among all methods, multi-atlas based segmentation methods are widely used, due to their robustness in propagating prior label information. However, non-linear registration is always needed, which is time-consuming. Alternatively, the patch-based methods have been proposed to relax the requirement of image registration, but the labeling is often determined independently by the target image information, without getting direct assistance from the atlases. To address these limitations, in this paper, we propose a multi-atlas guided 3D fully convolutional networks (FCN) for brain image labeling. Specifically, multi-atlas based guidance is incorporated during the network learning. Based on this, the discriminative of the FCN is boosted, which eventually contribute to accurate prediction. Experiments show that the use of multi-atlas guidance improves the brain labeling performance.

AB - Automatic labeling of anatomical structures in brain images plays an important role in neuroimaging analysis. Among all methods, multi-atlas based segmentation methods are widely used, due to their robustness in propagating prior label information. However, non-linear registration is always needed, which is time-consuming. Alternatively, the patch-based methods have been proposed to relax the requirement of image registration, but the labeling is often determined independently by the target image information, without getting direct assistance from the atlases. To address these limitations, in this paper, we propose a multi-atlas guided 3D fully convolutional networks (FCN) for brain image labeling. Specifically, multi-atlas based guidance is incorporated during the network learning. Based on this, the discriminative of the FCN is boosted, which eventually contribute to accurate prediction. Experiments show that the use of multi-atlas guidance improves the brain labeling performance.

UR - http://www.scopus.com/inward/record.url?scp=85029425257&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029425257&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-67434-6_2

DO - 10.1007/978-3-319-67434-6_2

M3 - Conference contribution

AN - SCOPUS:85029425257

SN - 9783319674339

VL - 10530 LNCS

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

SP - 12

EP - 19

BT - Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings

PB - Springer Verlag

ER -