Feature learning with component selective encoding for histopathology image classification

Yang Song; Hang Chang; Yang Gao; Sidong Liu; Donghao Zhang; Junen Yao; Wojciech Chrzanowski; Weidong Cai

doi:10.1109/ISBI.2018.8363568

Feature learning with component selective encoding for histopathology image classification

Yang Song
, Hang Chang
, Yang Gao
, Sidong Liu
, Donghao Zhang
, Junen Yao
, Wojciech Chrzanowski
, Weidong Cai

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

19 Scopus citations

Abstract

In this paper, we present a new feature representation method, called the Component Selective Encoding (CSE), for automated histopathology image classification. While the integration of Fisher Vector (FV) encoding with convolutional neural network (CNN) has demonstrated excellent performance in the classification of both general texture and histopathology images, the high dimensionality of FV descriptors could lead to suboptimal performance. Our proposed CSE method provides effective dimensionality reduction that is adaptive to the discriminativeness of individual Gaussian components in the FV descriptors. Evaluation on the publicly available BreaKHis dataset shows that our method outperforms the existing approaches based on deep learning and FV encoding.

Original language	English
Title of host publication	2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
Publisher	IEEE Computer Society
Pages	257-260
Number of pages	4
ISBN (Electronic)	9781538636367
DOIs	https://doi.org/10.1109/ISBI.2018.8363568
State	Published - 23 May 2018
Event	15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 - Washington, United States Duration: 4 Apr 2018 → 7 Apr 2018

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
Volume	2018-April
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
Country/Territory	United States
City	Washington
Period	4/04/18 → 7/04/18

Keywords

Dimensionality reduction
Fisher Vector
Histopathology images
Transfer learning

Access to Document

10.1109/ISBI.2018.8363568

Cite this

Song, Y., Chang, H., Gao, Y., Liu, S., Zhang, D., Yao, J., Chrzanowski, W., & Cai, W. (2018). Feature learning with component selective encoding for histopathology image classification. In 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018 (pp. 257-260). (Proceedings - International Symposium on Biomedical Imaging; Vol. 2018-April). IEEE Computer Society. https://doi.org/10.1109/ISBI.2018.8363568

@inproceedings{6d230be01b344208bbb905347a6ddb46,

title = "Feature learning with component selective encoding for histopathology image classification",

abstract = "In this paper, we present a new feature representation method, called the Component Selective Encoding (CSE), for automated histopathology image classification. While the integration of Fisher Vector (FV) encoding with convolutional neural network (CNN) has demonstrated excellent performance in the classification of both general texture and histopathology images, the high dimensionality of FV descriptors could lead to suboptimal performance. Our proposed CSE method provides effective dimensionality reduction that is adaptive to the discriminativeness of individual Gaussian components in the FV descriptors. Evaluation on the publicly available BreaKHis dataset shows that our method outperforms the existing approaches based on deep learning and FV encoding.",

keywords = "Dimensionality reduction, Fisher Vector, Histopathology images, Transfer learning",

author = "Yang Song and Hang Chang and Yang Gao and Sidong Liu and Donghao Zhang and Junen Yao and Wojciech Chrzanowski and Weidong Cai",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 ; Conference date: 04-04-2018 Through 07-04-2018",

year = "2018",

month = may,

day = "23",

doi = "10.1109/ISBI.2018.8363568",

language = "英语",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

pages = "257--260",

booktitle = "2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018",

address = "美国",

}

Song, Y, Chang, H, Gao, Y, Liu, S, Zhang, D, Yao, J, Chrzanowski, W & Cai, W 2018, Feature learning with component selective encoding for histopathology image classification. in 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. Proceedings - International Symposium on Biomedical Imaging, vol. 2018-April, IEEE Computer Society, pp. 257-260, 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018, Washington, United States, 4/04/18. https://doi.org/10.1109/ISBI.2018.8363568

Feature learning with component selective encoding for histopathology image classification. / Song, Yang; Chang, Hang; Gao, Yang et al.
2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. IEEE Computer Society, 2018. p. 257-260 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2018-April).

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

TY - GEN

T1 - Feature learning with component selective encoding for histopathology image classification

AU - Song, Yang

AU - Chang, Hang

AU - Gao, Yang

AU - Liu, Sidong

AU - Zhang, Donghao

AU - Yao, Junen

AU - Chrzanowski, Wojciech

AU - Cai, Weidong

PY - 2018/5/23

Y1 - 2018/5/23

N2 - In this paper, we present a new feature representation method, called the Component Selective Encoding (CSE), for automated histopathology image classification. While the integration of Fisher Vector (FV) encoding with convolutional neural network (CNN) has demonstrated excellent performance in the classification of both general texture and histopathology images, the high dimensionality of FV descriptors could lead to suboptimal performance. Our proposed CSE method provides effective dimensionality reduction that is adaptive to the discriminativeness of individual Gaussian components in the FV descriptors. Evaluation on the publicly available BreaKHis dataset shows that our method outperforms the existing approaches based on deep learning and FV encoding.

AB - In this paper, we present a new feature representation method, called the Component Selective Encoding (CSE), for automated histopathology image classification. While the integration of Fisher Vector (FV) encoding with convolutional neural network (CNN) has demonstrated excellent performance in the classification of both general texture and histopathology images, the high dimensionality of FV descriptors could lead to suboptimal performance. Our proposed CSE method provides effective dimensionality reduction that is adaptive to the discriminativeness of individual Gaussian components in the FV descriptors. Evaluation on the publicly available BreaKHis dataset shows that our method outperforms the existing approaches based on deep learning and FV encoding.

KW - Dimensionality reduction

KW - Fisher Vector

KW - Histopathology images

KW - Transfer learning

UR - https://www.scopus.com/pages/publications/85048073208

U2 - 10.1109/ISBI.2018.8363568

DO - 10.1109/ISBI.2018.8363568

M3 - 会议稿件

AN - SCOPUS:85048073208

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 257

EP - 260

BT - 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018

PB - IEEE Computer Society

T2 - 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018

Y2 - 4 April 2018 through 7 April 2018

ER -

Feature learning with component selective encoding for histopathology image classification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this