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)
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
UR - http://www.scopus.com/inward/citedby.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 -