A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT

Zunjie Xiao; Xiaoqing Zhang; Qingyang Sun; Zhuofei Wei; Gelei Xu; Yuan Jin; Risa Higashita; Jiang Liu

doi:10.1007/978-3-031-18910-4_22

A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT

Zunjie Xiao, Xiaoqing Zhang, Qingyang Sun, Zhuofei Wei, Gelei Xu, Yuan Jin, Risa Higashita, Jiang Liu

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

2 Citations (Scopus)

Abstract

Cataracts are the leading causes of reversible blindness and visual impairment. Cortical cataract (CC) is one of the most common cataract types. Early intervention is a useful way to improve the vision of CC patients. Anterior segment optical coherence tomography (AS-OCT) is a new ophthalmic image type that can capture the cortex region clearly. To our best knowledge, no previous work has paid attention to AS-OCT image-based CC classification. Thus, this paper presents an effective local-global spatial fusion network (LGSF-Net) to automatically recognize the severity levels of CC. In the LGSF-Net, we construct a novel local-global spatial fusion (LGSF) attention block by incorporating the clinical prior knowledge of CC. The extensive experiments conduct on the CASIA2 AS-OCT dataset of CC, and the results show that our LGSF-Net achieves better performance than state-of-the-art attention-based convolutional neural networks (CNNs). Moreover, we also demonstrate the generalization of our method on the public ChestXRay2017 dataset through comparisons to existing methods.

Original language	English
Title of host publication	Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings
Editors	Shiqi Yu, Jianguo Zhang, Zhaoxiang Zhang, Tieniu Tan, Pong C. Yuen, Yike Guo, Junwei Han, Jianhuang Lai
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	262-273
Number of pages	12
ISBN (Print)	9783031189098
DOIs	https://doi.org/10.1007/978-3-031-18910-4_22
Publication status	Published - 2022
Externally published	Yes
Event	5th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2022 - Shenzhen, China Duration: 4 Nov 2022 → 7 Nov 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13535 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2022
Country/Territory	China
City	Shenzhen
Period	4/11/22 → 7/11/22

Keywords

AS-OCT
Cortical cataract classification
Local-global

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-031-18910-4_22

Cite this

Xiao, Z., Zhang, X., Sun, Q., Wei, Z., Xu, G., Jin, Y., Higashita, R., & Liu, J. (2022). A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT. In S. Yu, J. Zhang, Z. Zhang, T. Tan, P. C. Yuen, Y. Guo, J. Han, & J. Lai (Eds.), Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings (pp. 262-273). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13535 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-18910-4_22

Xiao, Zunjie ; Zhang, Xiaoqing ; Sun, Qingyang et al. / A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT. Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings. editor / Shiqi Yu ; Jianguo Zhang ; Zhaoxiang Zhang ; Tieniu Tan ; Pong C. Yuen ; Yike Guo ; Junwei Han ; Jianhuang Lai. Springer Science and Business Media Deutschland GmbH, 2022. pp. 262-273 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{e33464bbcbcd407f87df058db64d6973,

title = "A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT",

abstract = "Cataracts are the leading causes of reversible blindness and visual impairment. Cortical cataract (CC) is one of the most common cataract types. Early intervention is a useful way to improve the vision of CC patients. Anterior segment optical coherence tomography (AS-OCT) is a new ophthalmic image type that can capture the cortex region clearly. To our best knowledge, no previous work has paid attention to AS-OCT image-based CC classification. Thus, this paper presents an effective local-global spatial fusion network (LGSF-Net) to automatically recognize the severity levels of CC. In the LGSF-Net, we construct a novel local-global spatial fusion (LGSF) attention block by incorporating the clinical prior knowledge of CC. The extensive experiments conduct on the CASIA2 AS-OCT dataset of CC, and the results show that our LGSF-Net achieves better performance than state-of-the-art attention-based convolutional neural networks (CNNs). Moreover, we also demonstrate the generalization of our method on the public ChestXRay2017 dataset through comparisons to existing methods.",

keywords = "AS-OCT, Cortical cataract classification, Local-global",

author = "Zunjie Xiao and Xiaoqing Zhang and Qingyang Sun and Zhuofei Wei and Gelei Xu and Yuan Jin and Risa Higashita and Jiang Liu",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 5th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2022 ; Conference date: 04-11-2022 Through 07-11-2022",

year = "2022",

doi = "10.1007/978-3-031-18910-4_22",

language = "English",

isbn = "9783031189098",

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

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "262--273",

editor = "Shiqi Yu and Jianguo Zhang and Zhaoxiang Zhang and Tieniu Tan and Yuen, {Pong C.} and Yike Guo and Junwei Han and Jianhuang Lai",

booktitle = "Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings",

address = "Germany",

}

Xiao, Z, Zhang, X, Sun, Q, Wei, Z, Xu, G, Jin, Y, Higashita, R & Liu, J 2022, A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT. in S Yu, J Zhang, Z Zhang, T Tan, PC Yuen, Y Guo, J Han & J Lai (eds), Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13535 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 262-273, 5th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2022, Shenzhen, China, 4/11/22. https://doi.org/10.1007/978-3-031-18910-4_22

A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT. / Xiao, Zunjie; Zhang, Xiaoqing; Sun, Qingyang et al.
Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings. ed. / Shiqi Yu; Jianguo Zhang; Zhaoxiang Zhang; Tieniu Tan; Pong C. Yuen; Yike Guo; Junwei Han; Jianhuang Lai. Springer Science and Business Media Deutschland GmbH, 2022. p. 262-273 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13535 LNCS).

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

TY - GEN

T1 - A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT

AU - Xiao, Zunjie

AU - Zhang, Xiaoqing

AU - Sun, Qingyang

AU - Wei, Zhuofei

AU - Xu, Gelei

AU - Jin, Yuan

AU - Higashita, Risa

AU - Liu, Jiang

PY - 2022

Y1 - 2022

N2 - Cataracts are the leading causes of reversible blindness and visual impairment. Cortical cataract (CC) is one of the most common cataract types. Early intervention is a useful way to improve the vision of CC patients. Anterior segment optical coherence tomography (AS-OCT) is a new ophthalmic image type that can capture the cortex region clearly. To our best knowledge, no previous work has paid attention to AS-OCT image-based CC classification. Thus, this paper presents an effective local-global spatial fusion network (LGSF-Net) to automatically recognize the severity levels of CC. In the LGSF-Net, we construct a novel local-global spatial fusion (LGSF) attention block by incorporating the clinical prior knowledge of CC. The extensive experiments conduct on the CASIA2 AS-OCT dataset of CC, and the results show that our LGSF-Net achieves better performance than state-of-the-art attention-based convolutional neural networks (CNNs). Moreover, we also demonstrate the generalization of our method on the public ChestXRay2017 dataset through comparisons to existing methods.

AB - Cataracts are the leading causes of reversible blindness and visual impairment. Cortical cataract (CC) is one of the most common cataract types. Early intervention is a useful way to improve the vision of CC patients. Anterior segment optical coherence tomography (AS-OCT) is a new ophthalmic image type that can capture the cortex region clearly. To our best knowledge, no previous work has paid attention to AS-OCT image-based CC classification. Thus, this paper presents an effective local-global spatial fusion network (LGSF-Net) to automatically recognize the severity levels of CC. In the LGSF-Net, we construct a novel local-global spatial fusion (LGSF) attention block by incorporating the clinical prior knowledge of CC. The extensive experiments conduct on the CASIA2 AS-OCT dataset of CC, and the results show that our LGSF-Net achieves better performance than state-of-the-art attention-based convolutional neural networks (CNNs). Moreover, we also demonstrate the generalization of our method on the public ChestXRay2017 dataset through comparisons to existing methods.

KW - AS-OCT

KW - Cortical cataract classification

KW - Local-global

UR - http://www.scopus.com/inward/record.url?scp=85142746204&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-18910-4_22

DO - 10.1007/978-3-031-18910-4_22

M3 - Conference contribution

AN - SCOPUS:85142746204

SN - 9783031189098

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

SP - 262

EP - 273

BT - Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings

A2 - Yu, Shiqi

A2 - Zhang, Jianguo

A2 - Zhang, Zhaoxiang

A2 - Tan, Tieniu

A2 - Yuen, Pong C.

A2 - Guo, Yike

A2 - Han, Junwei

A2 - Lai, Jianhuang

PB - Springer Science and Business Media Deutschland GmbH

T2 - 5th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2022

Y2 - 4 November 2022 through 7 November 2022

ER -

Xiao Z, Zhang X, Sun Q, Wei Z, Xu G, Jin Y et al. A Novel Local-Global Spatial Attention Network for Cortical Cataract Classification in AS-OCT. In Yu S, Zhang J, Zhang Z, Tan T, Yuen PC, Guo Y, Han J, Lai J, editors, Pattern Recognition and Computer Vision - 5th Chinese Conference, PRCV 2022, Proceedings. Springer Science and Business Media Deutschland GmbH. 2022. p. 262-273. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-18910-4_22