A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis

Shujiao Sun; Bonan Jiang; Yushan Zheng; Fengying Xie

doi:10.1007/978-3-030-34110-7_47

A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis

Shujiao Sun
, Bonan Jiang
, Yushan Zheng^*
, Fengying Xie

^*此作品的通讯作者

Beihang University
Beijing University of Technology

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

4 引用（Scopus）

摘要

Automatic analysis of histopathological whole slide images (WSIs) is a challenging task. In this paper, we designed two deep learning structures based on a fully convolutional network (FCN) and a convolutional neural network (CNN), to achieve the segmentation of carcinoma regions from WSIs. FCN is developed for segmentation problems and CNN focuses on classification. We designed experiments to compare the performances of the two methods. The results demonstrated that CNN performs as well as FCN when applied to WSIs in high resolution. Furthermore, to leverage the advantages of CNN and FCN, we integrate the two methods to obtain a complete framework for lung cancer segmentation. The proposed methods were evaluated on the ACDC-LungHP dataset. The final dice coefficient for cancerous region segmentation is 0.770.

源语言	英语
主期刊名	Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2
编辑	Yao Zhao, Chunyu Lin, Nick Barnes, Baoquan Chen, Rüdiger Westermann, Xiangwei Kong
出版商	Springer
页	558-567
页数	10
ISBN（印刷版）	9783030341091
DOI	https://doi.org/10.1007/978-3-030-34110-7_47
出版状态	已出版 - 2019
活动	10th International Conference on Image and Graphics, ICIG 2019 - Beijing, 中国期限: 23 8月 2019 → 25 8月 2019

出版系列

姓名	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
卷	11902 LNCS
ISSN（印刷版）	0302-9743
ISSN（电子版）	1611-3349

会议

会议	10th International Conference on Image and Graphics, ICIG 2019
国家/地区	中国
市	Beijing
时期	23/08/19 → 25/08/19

联合国可持续发展目标

此成果有助于实现下列可持续发展目标：

可持续发展目标 3 良好健康与福祉

访问文件

10.1007/978-3-030-34110-7_47

其它文件与链接

链接到 Scopus 的出版物

引用此

Sun, S., Jiang, B., Zheng, Y., & Xie, F. (2019). A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis. 在 Y. Zhao, C. Lin, N. Barnes, B. Chen, R. Westermann, & X. Kong (编辑), Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2 (页码 558-567). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 卷 11902 LNCS). Springer. https://doi.org/10.1007/978-3-030-34110-7_47

Sun, Shujiao ; Jiang, Bonan ; Zheng, Yushan 等. / A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis. Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2. 编辑 / Yao Zhao ; Chunyu Lin ; Nick Barnes ; Baoquan Chen ; Rüdiger Westermann ; Xiangwei Kong. Springer, 2019. 页码 558-567 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{59a04d67ca1041d6adc5f08816c8006d,

title = "A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis",

abstract = "Automatic analysis of histopathological whole slide images (WSIs) is a challenging task. In this paper, we designed two deep learning structures based on a fully convolutional network (FCN) and a convolutional neural network (CNN), to achieve the segmentation of carcinoma regions from WSIs. FCN is developed for segmentation problems and CNN focuses on classification. We designed experiments to compare the performances of the two methods. The results demonstrated that CNN performs as well as FCN when applied to WSIs in high resolution. Furthermore, to leverage the advantages of CNN and FCN, we integrate the two methods to obtain a complete framework for lung cancer segmentation. The proposed methods were evaluated on the ACDC-LungHP dataset. The final dice coefficient for cancerous region segmentation is 0.770.",

keywords = "CNN, Computational pathology, FCN, Image segmentation, Lung cancer",

author = "Shujiao Sun and Bonan Jiang and Yushan Zheng and Fengying Xie",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Nature Switzerland AG.; 10th International Conference on Image and Graphics, ICIG 2019 ; Conference date: 23-08-2019 Through 25-08-2019",

year = "2019",

doi = "10.1007/978-3-030-34110-7\_47",

language = "英语",

isbn = "9783030341091",

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

publisher = "Springer",

pages = "558--567",

editor = "Yao Zhao and Chunyu Lin and Nick Barnes and Baoquan Chen and R{\"u}diger Westermann and Xiangwei Kong",

booktitle = "Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2",

address = "德国",

}

Sun, S, Jiang, B, Zheng, Y & Xie, F 2019, A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis. 在 Y Zhao, C Lin, N Barnes, B Chen, R Westermann & X Kong (编辑), Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 卷 11902 LNCS, Springer, 页码 558-567, 10th International Conference on Image and Graphics, ICIG 2019, Beijing, 中国, 23/08/19. https://doi.org/10.1007/978-3-030-34110-7_47

A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis. / Sun, Shujiao; Jiang, Bonan; Zheng, Yushan 等.
Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2. 编辑 / Yao Zhao; Chunyu Lin; Nick Barnes; Baoquan Chen; Rüdiger Westermann; Xiangwei Kong. Springer, 2019. 页码 558-567 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 卷 11902 LNCS).

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

TY - GEN

T1 - A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis

AU - Sun, Shujiao

AU - Jiang, Bonan

AU - Zheng, Yushan

AU - Xie, Fengying

PY - 2019

Y1 - 2019

N2 - Automatic analysis of histopathological whole slide images (WSIs) is a challenging task. In this paper, we designed two deep learning structures based on a fully convolutional network (FCN) and a convolutional neural network (CNN), to achieve the segmentation of carcinoma regions from WSIs. FCN is developed for segmentation problems and CNN focuses on classification. We designed experiments to compare the performances of the two methods. The results demonstrated that CNN performs as well as FCN when applied to WSIs in high resolution. Furthermore, to leverage the advantages of CNN and FCN, we integrate the two methods to obtain a complete framework for lung cancer segmentation. The proposed methods were evaluated on the ACDC-LungHP dataset. The final dice coefficient for cancerous region segmentation is 0.770.

AB - Automatic analysis of histopathological whole slide images (WSIs) is a challenging task. In this paper, we designed two deep learning structures based on a fully convolutional network (FCN) and a convolutional neural network (CNN), to achieve the segmentation of carcinoma regions from WSIs. FCN is developed for segmentation problems and CNN focuses on classification. We designed experiments to compare the performances of the two methods. The results demonstrated that CNN performs as well as FCN when applied to WSIs in high resolution. Furthermore, to leverage the advantages of CNN and FCN, we integrate the two methods to obtain a complete framework for lung cancer segmentation. The proposed methods were evaluated on the ACDC-LungHP dataset. The final dice coefficient for cancerous region segmentation is 0.770.

KW - CNN

KW - Computational pathology

KW - FCN

KW - Image segmentation

KW - Lung cancer

UR - https://www.scopus.com/pages/publications/85076922930

U2 - 10.1007/978-3-030-34110-7_47

DO - 10.1007/978-3-030-34110-7_47

M3 - 会议稿件

AN - SCOPUS:85076922930

SN - 9783030341091

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

SP - 558

EP - 567

BT - Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2

A2 - Zhao, Yao

A2 - Lin, Chunyu

A2 - Barnes, Nick

A2 - Chen, Baoquan

A2 - Westermann, Rüdiger

A2 - Kong, Xiangwei

PB - Springer

T2 - 10th International Conference on Image and Graphics, ICIG 2019

Y2 - 23 August 2019 through 25 August 2019

ER -

Sun S, Jiang B, Zheng Y , Xie F. A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis. 在 Zhao Y, Lin C, Barnes N, Chen B, Westermann R, Kong X, 编辑, Image and Graphics - 10th International Conference, ICIG 2019, Proceedings, Part 2. Springer. 2019. 页码 558-567. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-34110-7_47

A Comparative Study of CNN and FCN for Histopathology Whole Slide Image Analysis

摘要

出版系列

会议

联合国可持续发展目标

访问文件

其它文件与链接

学术指纹

引用此