Multiscale network followed network model for retinal vessel segmentation

Yicheng Wu; Yong Xia; Yang Song; Yanning Zhang; Weidong Cai

doi:10.1007/978-3-030-00934-2_14

Multiscale network followed network model for retinal vessel segmentation

Yicheng Wu
, Yong Xia
, Yang Song
, Yanning Zhang
, Weidong Cai

School of Computer Science

Northwestern Polytechnical University Xian
The University of Sydney

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

169 Scopus citations

Abstract

The shape of retinal blood vessels plays a critical role in the early diagnosis of diabetic retinopathy. However, it remains challenging to segment accurately the blood vessels, particularly the capillaries, in color retinal images. In this paper, we propose the multiscale network followed network (MS-NFN) model to address this issue. This model consists of an ‘up-pool’ NFN submodel and a ‘pool-up’ NFN submodel, in which max-pooling layers and up-sampling layers can generate multiscale feature maps. In each NFN, the first multiscale network converts an image patch into a probabilistic retinal vessel map, and the following multiscale network further refines the map. The refined probabilistic retinal vessel maps produced by both NFNs are averaged to construct the segmentation result. We evaluated this model on the digital retinal images for vessel extraction (DRIVE) dataset and the child heart and health study dataset. Our results indicate that the NFN structure we designed is able to produce performance gain and the proposed MS-NFN model achieved the state-of-the-art retinal vessel segmentation accuracy on both datasets.

Original language	English
Title of host publication	Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings
Editors	Gabor Fichtinger, Christos Davatzikos, Carlos Alberola-López, Alejandro F. Frangi, Julia A. Schnabel
Publisher	Springer Verlag
Pages	119-126
Number of pages	8
ISBN (Print)	9783030009335
DOIs	https://doi.org/10.1007/978-3-030-00934-2_14
State	Published - 2018
Event	21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 - Granada, Spain Duration: 16 Sep 2018 → 20 Sep 2018

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11071 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018
Country/Territory	Spain
City	Granada
Period	16/09/18 → 20/09/18

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Fully convolutional network
Network followed network
Retinal vessel segmentation

Access to Document

10.1007/978-3-030-00934-2_14

Cite this

Wu, Y., Xia, Y., Song, Y., Zhang, Y., & Cai, W. (2018). Multiscale network followed network model for retinal vessel segmentation. In G. Fichtinger, C. Davatzikos, C. Alberola-López, A. F. Frangi, & J. A. Schnabel (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings (pp. 119-126). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11071 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-00934-2_14

Wu, Yicheng ; Xia, Yong ; Song, Yang et al. / Multiscale network followed network model for retinal vessel segmentation. Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. editor / Gabor Fichtinger ; Christos Davatzikos ; Carlos Alberola-López ; Alejandro F. Frangi ; Julia A. Schnabel. Springer Verlag, 2018. pp. 119-126 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ad749ec88bd8460aa676782cba698204,

title = "Multiscale network followed network model for retinal vessel segmentation",

abstract = "The shape of retinal blood vessels plays a critical role in the early diagnosis of diabetic retinopathy. However, it remains challenging to segment accurately the blood vessels, particularly the capillaries, in color retinal images. In this paper, we propose the multiscale network followed network (MS-NFN) model to address this issue. This model consists of an {\textquoteleft}up-pool{\textquoteright} NFN submodel and a {\textquoteleft}pool-up{\textquoteright} NFN submodel, in which max-pooling layers and up-sampling layers can generate multiscale feature maps. In each NFN, the first multiscale network converts an image patch into a probabilistic retinal vessel map, and the following multiscale network further refines the map. The refined probabilistic retinal vessel maps produced by both NFNs are averaged to construct the segmentation result. We evaluated this model on the digital retinal images for vessel extraction (DRIVE) dataset and the child heart and health study dataset. Our results indicate that the NFN structure we designed is able to produce performance gain and the proposed MS-NFN model achieved the state-of-the-art retinal vessel segmentation accuracy on both datasets.",

keywords = "Fully convolutional network, Network followed network, Retinal vessel segmentation",

author = "Yicheng Wu and Yong Xia and Yang Song and Yanning Zhang and Weidong Cai",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2018.; 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018 ; Conference date: 16-09-2018 Through 20-09-2018",

year = "2018",

doi = "10.1007/978-3-030-00934-2\_14",

language = "英语",

isbn = "9783030009335",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "119--126",

editor = "Gabor Fichtinger and Christos Davatzikos and Carlos Alberola-L{\'o}pez and Frangi, \{Alejandro F.\} and Schnabel, \{Julia A.\}",

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

}

Wu, Y, Xia, Y, Song, Y, Zhang, Y & Cai, W 2018, Multiscale network followed network model for retinal vessel segmentation. in G Fichtinger, C Davatzikos, C Alberola-López, AF Frangi & JA Schnabel (eds), Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11071 LNCS, Springer Verlag, pp. 119-126, 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018, Granada, Spain, 16/09/18. https://doi.org/10.1007/978-3-030-00934-2_14

Multiscale network followed network model for retinal vessel segmentation. / Wu, Yicheng; Xia, Yong; Song, Yang et al.
Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. ed. / Gabor Fichtinger; Christos Davatzikos; Carlos Alberola-López; Alejandro F. Frangi; Julia A. Schnabel. Springer Verlag, 2018. p. 119-126 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11071 LNCS).

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

TY - GEN

T1 - Multiscale network followed network model for retinal vessel segmentation

AU - Wu, Yicheng

AU - Xia, Yong

AU - Song, Yang

AU - Zhang, Yanning

AU - Cai, Weidong

N1 - Publisher Copyright: © Springer Nature Switzerland AG 2018.

PY - 2018

Y1 - 2018

N2 - The shape of retinal blood vessels plays a critical role in the early diagnosis of diabetic retinopathy. However, it remains challenging to segment accurately the blood vessels, particularly the capillaries, in color retinal images. In this paper, we propose the multiscale network followed network (MS-NFN) model to address this issue. This model consists of an ‘up-pool’ NFN submodel and a ‘pool-up’ NFN submodel, in which max-pooling layers and up-sampling layers can generate multiscale feature maps. In each NFN, the first multiscale network converts an image patch into a probabilistic retinal vessel map, and the following multiscale network further refines the map. The refined probabilistic retinal vessel maps produced by both NFNs are averaged to construct the segmentation result. We evaluated this model on the digital retinal images for vessel extraction (DRIVE) dataset and the child heart and health study dataset. Our results indicate that the NFN structure we designed is able to produce performance gain and the proposed MS-NFN model achieved the state-of-the-art retinal vessel segmentation accuracy on both datasets.

AB - The shape of retinal blood vessels plays a critical role in the early diagnosis of diabetic retinopathy. However, it remains challenging to segment accurately the blood vessels, particularly the capillaries, in color retinal images. In this paper, we propose the multiscale network followed network (MS-NFN) model to address this issue. This model consists of an ‘up-pool’ NFN submodel and a ‘pool-up’ NFN submodel, in which max-pooling layers and up-sampling layers can generate multiscale feature maps. In each NFN, the first multiscale network converts an image patch into a probabilistic retinal vessel map, and the following multiscale network further refines the map. The refined probabilistic retinal vessel maps produced by both NFNs are averaged to construct the segmentation result. We evaluated this model on the digital retinal images for vessel extraction (DRIVE) dataset and the child heart and health study dataset. Our results indicate that the NFN structure we designed is able to produce performance gain and the proposed MS-NFN model achieved the state-of-the-art retinal vessel segmentation accuracy on both datasets.

KW - Fully convolutional network

KW - Network followed network

KW - Retinal vessel segmentation

UR - https://www.scopus.com/pages/publications/85054074994

U2 - 10.1007/978-3-030-00934-2_14

DO - 10.1007/978-3-030-00934-2_14

M3 - 会议稿件

AN - SCOPUS:85054074994

SN - 9783030009335

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

SP - 119

EP - 126

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

A2 - Fichtinger, Gabor

A2 - Davatzikos, Christos

A2 - Alberola-López, Carlos

A2 - Frangi, Alejandro F.

A2 - Schnabel, Julia A.

PB - Springer Verlag

T2 - 21st International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2018

Y2 - 16 September 2018 through 20 September 2018

ER -

Wu Y, Xia Y, Song Y, Zhang Y, Cai W. Multiscale network followed network model for retinal vessel segmentation. In Fichtinger G, Davatzikos C, Alberola-López C, Frangi AF, Schnabel JA, editors, Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 - 21st International Conference, 2018, Proceedings. Springer Verlag. 2018. p. 119-126. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-00934-2_14

Multiscale network followed network model for retinal vessel segmentation

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this