Self-attentive normalization for automated gleason grading system

Hong Kyu Shin; Sung Hoo Hong; Yeong Jin Choi; Yong Goo Shin; Seung Park; Sung Jea Ko

doi:10.1109/TENCON50793.2020.9293775

Self-attentive normalization for automated gleason grading system

Hong Kyu Shin, Sung Hoo Hong, Yeong Jin Choi, Yong Goo Shin, Seung Park, Sung Jea Ko

School of Electrical Engineering

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

Abstract

Recently, convolutional neural networks (CNNs)- based automated Gleason grading system for prostate cancer has been widely researched. However, these systems still need further improvement to achieve pathologist-level performance. To this end, this paper introduces a novel self-attentive normalization (SAN) which is the first work to employ the attention mechanism for the automated Gleason grading system. Unlike conventional normalization techniques, e.g. batch normalization and instance normalization, which learn a single affine transformation, the proposed method can learn the elementwise affine transformation to focus on more informative regions of the feature map. Since SAN requires a small number of extra learning parameters, it can be integrated into existing automated Gleason grading systems seamlessly with negligible overheads. Extensive quantitative evaluations show that, by applying SAN to various CNN architectures, the diagnostic accuracy can be significantly improved. For instance, we raise VGG-16's diagnostic accuracy from 73.99% to 79.16% on the Harvard Dataverse.

Original language	English
Title of host publication	2020 IEEE Region 10 Conference, TENCON 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1101-1105
Number of pages	5
ISBN (Electronic)	9781728184555
DOIs	https://doi.org/10.1109/TENCON50793.2020.9293775
Publication status	Published - 2020 Nov 16
Event	2020 IEEE Region 10 Conference, TENCON 2020 - Virtual, Osaka, Japan Duration: 2020 Nov 16 → 2020 Nov 19

Publication series

Name	IEEE Region 10 Annual International Conference, Proceedings/TENCON
Volume	2020-November
ISSN (Print)	2159-3442
ISSN (Electronic)	2159-3450

Conference

Conference	2020 IEEE Region 10 Conference, TENCON 2020
Country	Japan
City	Virtual, Osaka
Period	20/11/16 → 20/11/19

Keywords

Automated Gleason grading system
Convolutional neural networks
Self-attentive normalization

ASJC Scopus subject areas

Computer Science Applications
Electrical and Electronic Engineering

UN SDGs

This output contributes to the following Sustainable Development Goal(s)

Access to Document

10.1109/TENCON50793.2020.9293775

Fingerprint Dive into the research topics of 'Self-attentive normalization for automated gleason grading system'. Together they form a unique fingerprint.

Cite this

Shin, H. K., Hong, S. H., Choi, Y. J., Shin, Y. G., Park, S., & Ko, S. J. (2020). Self-attentive normalization for automated gleason grading system. In 2020 IEEE Region 10 Conference, TENCON 2020 (pp. 1101-1105). [9293775] (IEEE Region 10 Annual International Conference, Proceedings/TENCON; Vol. 2020-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/TENCON50793.2020.9293775

Self-attentive normalization for automated gleason grading system. / Shin, Hong Kyu; Hong, Sung Hoo; Choi, Yeong Jin; Shin, Yong Goo; Park, Seung; Ko, Sung Jea.

2020 IEEE Region 10 Conference, TENCON 2020. Institute of Electrical and Electronics Engineers Inc., 2020. p. 1101-1105 9293775 (IEEE Region 10 Annual International Conference, Proceedings/TENCON; Vol. 2020-November).

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

Shin, HK, Hong, SH, Choi, YJ, Shin, YG, Park, S & Ko, SJ 2020, Self-attentive normalization for automated gleason grading system. in 2020 IEEE Region 10 Conference, TENCON 2020., 9293775, IEEE Region 10 Annual International Conference, Proceedings/TENCON, vol. 2020-November, Institute of Electrical and Electronics Engineers Inc., pp. 1101-1105, 2020 IEEE Region 10 Conference, TENCON 2020, Virtual, Osaka, Japan, 20/11/16. https://doi.org/10.1109/TENCON50793.2020.9293775

@inproceedings{5dc37835ebd24bbd8fa0807a174920c8,

title = "Self-attentive normalization for automated gleason grading system",

abstract = "Recently, convolutional neural networks (CNNs)- based automated Gleason grading system for prostate cancer has been widely researched. However, these systems still need further improvement to achieve pathologist-level performance. To this end, this paper introduces a novel self-attentive normalization (SAN) which is the first work to employ the attention mechanism for the automated Gleason grading system. Unlike conventional normalization techniques, e.g. batch normalization and instance normalization, which learn a single affine transformation, the proposed method can learn the elementwise affine transformation to focus on more informative regions of the feature map. Since SAN requires a small number of extra learning parameters, it can be integrated into existing automated Gleason grading systems seamlessly with negligible overheads. Extensive quantitative evaluations show that, by applying SAN to various CNN architectures, the diagnostic accuracy can be significantly improved. For instance, we raise VGG-16's diagnostic accuracy from 73.99% to 79.16% on the Harvard Dataverse.",

keywords = "Automated Gleason grading system, Convolutional neural networks, Self-attentive normalization",

author = "Shin, {Hong Kyu} and Hong, {Sung Hoo} and Choi, {Yeong Jin} and Shin, {Yong Goo} and Seung Park and Ko, {Sung Jea}",

note = "Funding Information: This work was supported by Institute for Information & communications Technology Promotion (IITP) grant funded by the Korea government(MSIT) (No.2019-0-00268, Development of SW technology for recognition, judgment and path control algorithm verification simulation and dataset generation) Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE Region 10 Conference, TENCON 2020 ; Conference date: 16-11-2020 Through 19-11-2020",

year = "2020",

month = nov,

day = "16",

doi = "10.1109/TENCON50793.2020.9293775",

language = "English",

series = "IEEE Region 10 Annual International Conference, Proceedings/TENCON",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1101--1105",

booktitle = "2020 IEEE Region 10 Conference, TENCON 2020",

}

TY - GEN

T1 - Self-attentive normalization for automated gleason grading system

AU - Shin, Hong Kyu

AU - Hong, Sung Hoo

AU - Choi, Yeong Jin

AU - Shin, Yong Goo

AU - Park, Seung

AU - Ko, Sung Jea

N1 - Funding Information: This work was supported by Institute for Information & communications Technology Promotion (IITP) grant funded by the Korea government(MSIT) (No.2019-0-00268, Development of SW technology for recognition, judgment and path control algorithm verification simulation and dataset generation) Publisher Copyright: © 2020 IEEE.

PY - 2020/11/16

Y1 - 2020/11/16

N2 - Recently, convolutional neural networks (CNNs)- based automated Gleason grading system for prostate cancer has been widely researched. However, these systems still need further improvement to achieve pathologist-level performance. To this end, this paper introduces a novel self-attentive normalization (SAN) which is the first work to employ the attention mechanism for the automated Gleason grading system. Unlike conventional normalization techniques, e.g. batch normalization and instance normalization, which learn a single affine transformation, the proposed method can learn the elementwise affine transformation to focus on more informative regions of the feature map. Since SAN requires a small number of extra learning parameters, it can be integrated into existing automated Gleason grading systems seamlessly with negligible overheads. Extensive quantitative evaluations show that, by applying SAN to various CNN architectures, the diagnostic accuracy can be significantly improved. For instance, we raise VGG-16's diagnostic accuracy from 73.99% to 79.16% on the Harvard Dataverse.

AB - Recently, convolutional neural networks (CNNs)- based automated Gleason grading system for prostate cancer has been widely researched. However, these systems still need further improvement to achieve pathologist-level performance. To this end, this paper introduces a novel self-attentive normalization (SAN) which is the first work to employ the attention mechanism for the automated Gleason grading system. Unlike conventional normalization techniques, e.g. batch normalization and instance normalization, which learn a single affine transformation, the proposed method can learn the elementwise affine transformation to focus on more informative regions of the feature map. Since SAN requires a small number of extra learning parameters, it can be integrated into existing automated Gleason grading systems seamlessly with negligible overheads. Extensive quantitative evaluations show that, by applying SAN to various CNN architectures, the diagnostic accuracy can be significantly improved. For instance, we raise VGG-16's diagnostic accuracy from 73.99% to 79.16% on the Harvard Dataverse.

KW - Automated Gleason grading system

KW - Convolutional neural networks

KW - Self-attentive normalization

UR - http://www.scopus.com/inward/record.url?scp=85098932662&partnerID=8YFLogxK

U2 - 10.1109/TENCON50793.2020.9293775

DO - 10.1109/TENCON50793.2020.9293775

M3 - Conference contribution

AN - SCOPUS:85098932662

T3 - IEEE Region 10 Annual International Conference, Proceedings/TENCON

SP - 1101

EP - 1105

BT - 2020 IEEE Region 10 Conference, TENCON 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 IEEE Region 10 Conference, TENCON 2020

Y2 - 16 November 2020 through 19 November 2020

ER -

Self-attentive normalization for automated gleason grading system

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Cite this