Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification

Xiaoqing Guo; Yixuan Yuan

doi:10.1007/978-3-030-32239-7_33

Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification

Xiaoqing Guo
, Yixuan Yuan^*

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

55 Citations (Scopus)

Abstract

Accurate detection of abnormal regions in Wireless Capsule Endoscopy (WCE) images is crucial for early intestine cancer diagnosis and treatment, while it still remains challenging due to the relatively low contrasts and ambiguous boundaries between abnormalities and normal regions. Additionally, the huge intra-class variances, alone with the high degree of visual similarities shared by inter-class abnormalities prevent the network from robust classification. To tackle these dilemmas, we propose an Adaptive Abnormal-aware Attention Network (Triple ANet) with Adaptive Dense Block (ADB) and Abnormal-aware Attention Module (AAM) for automatic WCE image analysis. ADB is designed to assign one attention score for each dense connection in dense blocks and to enhance useful features, while AAM aims to adaptively adjust the respective field according to the abnormal regions and help pay attention to abnormalities. Moreover, we propose a novel Angular Contrastive loss (AC Loss) to reduce the intra-class variances and enlarge the inter-class differences effectively. Our methods achieved 89.41% overall accuracy and showed better performance compared with state-of-the-art WCE image classification methods. The source code is available at https://github.com/Guo-Xiaoqing/Triple-ANet.

Original language	English
Title of host publication	Medical Image Computing and Computer Assisted Intervention – MICCAI 2019
Subtitle of host publication	22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I
Editors	Dinggang Shen, Pew-Thian Yap, Tianming Liu, Terry M. Peters, Ali Khan, Lawrence H. Staib, Caroline Essert, Sean Zhou
Publisher	Springer Cham
Pages	293-301
Number of pages	9
Edition	1st
ISBN (Electronic)	9783030322397
ISBN (Print)	9783030322380
DOIs	https://doi.org/10.1007/978-3-030-32239-7_33
Publication status	Published - 10 Oct 2019
Event	22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China Duration: 13 Oct 2019 → 17 Oct 2019

Publication series

Name	Lecture Notes in Computer Science
Volume	11764
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Country/Territory	China
City	Shenzhen
Period	13/10/19 → 17/10/19

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.1007/978-3-030-32239-7_33

Cite this

Guo, X., & Yuan, Y. (2019). Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification. In D. Shen, P.-T. Yap, T. Liu, T. M. Peters, A. Khan, L. H. Staib, C. Essert, & S. Zhou (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I (1st ed., pp. 293-301). (Lecture Notes in Computer Science; Vol. 11764). Springer Cham. https://doi.org/10.1007/978-3-030-32239-7_33

Guo, Xiaoqing ; Yuan, Yixuan. / Triple ANet : Adaptive Abnormal-aware Attention Network for WCE Image Classification. Medical Image Computing and Computer Assisted Intervention – MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I. editor / Dinggang Shen ; Pew-Thian Yap ; Tianming Liu ; Terry M. Peters ; Ali Khan ; Lawrence H. Staib ; Caroline Essert ; Sean Zhou. 1st. ed. Springer Cham, 2019. pp. 293-301 (Lecture Notes in Computer Science).

@inproceedings{182e9550627f4dfc9b4cbc7332445f9c,

title = "Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification",

abstract = "Accurate detection of abnormal regions in Wireless Capsule Endoscopy (WCE) images is crucial for early intestine cancer diagnosis and treatment, while it still remains challenging due to the relatively low contrasts and ambiguous boundaries between abnormalities and normal regions. Additionally, the huge intra-class variances, alone with the high degree of visual similarities shared by inter-class abnormalities prevent the network from robust classification. To tackle these dilemmas, we propose an Adaptive Abnormal-aware Attention Network (Triple ANet) with Adaptive Dense Block (ADB) and Abnormal-aware Attention Module (AAM) for automatic WCE image analysis. ADB is designed to assign one attention score for each dense connection in dense blocks and to enhance useful features, while AAM aims to adaptively adjust the respective field according to the abnormal regions and help pay attention to abnormalities. Moreover, we propose a novel Angular Contrastive loss (AC Loss) to reduce the intra-class variances and enlarge the inter-class differences effectively. Our methods achieved 89.41\% overall accuracy and showed better performance compared with state-of-the-art WCE image classification methods. The source code is available at https://github.com/Guo-Xiaoqing/Triple-ANet.",

author = "Xiaoqing Guo and Yixuan Yuan",

note = "Funding Information: This work was supported by Sichuan Provincial Science and Technology Research Grant 2019YJ0632 and TSG 6000690. We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan Xp GPU for this research. Publisher Copyright: {\textcopyright} 2019 Springer Nature Switzerland AG; 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 ; Conference date: 13-10-2019 Through 17-10-2019",

year = "2019",

month = oct,

day = "10",

doi = "10.1007/978-3-030-32239-7\_33",

language = "English",

isbn = "9783030322380",

series = "Lecture Notes in Computer Science",

publisher = "Springer Cham",

pages = "293--301",

editor = "Dinggang Shen and Pew-Thian Yap and Tianming Liu and Peters, \{Terry M.\} and Ali Khan and Staib, \{Lawrence H.\} and Caroline Essert and Sean Zhou",

booktitle = "Medical Image Computing and Computer Assisted Intervention – MICCAI 2019",

edition = "1st",

}

Guo, X & Yuan, Y 2019, Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification. in D Shen, P-T Yap, T Liu, TM Peters, A Khan, LH Staib, C Essert & S Zhou (eds), Medical Image Computing and Computer Assisted Intervention – MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I. 1st edn, Lecture Notes in Computer Science, vol. 11764, Springer Cham, pp. 293-301, 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019, Shenzhen, China, 13/10/19. https://doi.org/10.1007/978-3-030-32239-7_33

Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification. / Guo, Xiaoqing; Yuan, Yixuan.
Medical Image Computing and Computer Assisted Intervention – MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I. ed. / Dinggang Shen; Pew-Thian Yap; Tianming Liu; Terry M. Peters; Ali Khan; Lawrence H. Staib; Caroline Essert; Sean Zhou. 1st. ed. Springer Cham, 2019. p. 293-301 (Lecture Notes in Computer Science; Vol. 11764).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Triple ANet

T2 - 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019

AU - Guo, Xiaoqing

AU - Yuan, Yixuan

N1 - Funding Information: This work was supported by Sichuan Provincial Science and Technology Research Grant 2019YJ0632 and TSG 6000690. We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan Xp GPU for this research. Publisher Copyright: © 2019 Springer Nature Switzerland AG

PY - 2019/10/10

Y1 - 2019/10/10

N2 - Accurate detection of abnormal regions in Wireless Capsule Endoscopy (WCE) images is crucial for early intestine cancer diagnosis and treatment, while it still remains challenging due to the relatively low contrasts and ambiguous boundaries between abnormalities and normal regions. Additionally, the huge intra-class variances, alone with the high degree of visual similarities shared by inter-class abnormalities prevent the network from robust classification. To tackle these dilemmas, we propose an Adaptive Abnormal-aware Attention Network (Triple ANet) with Adaptive Dense Block (ADB) and Abnormal-aware Attention Module (AAM) for automatic WCE image analysis. ADB is designed to assign one attention score for each dense connection in dense blocks and to enhance useful features, while AAM aims to adaptively adjust the respective field according to the abnormal regions and help pay attention to abnormalities. Moreover, we propose a novel Angular Contrastive loss (AC Loss) to reduce the intra-class variances and enlarge the inter-class differences effectively. Our methods achieved 89.41% overall accuracy and showed better performance compared with state-of-the-art WCE image classification methods. The source code is available at https://github.com/Guo-Xiaoqing/Triple-ANet.

AB - Accurate detection of abnormal regions in Wireless Capsule Endoscopy (WCE) images is crucial for early intestine cancer diagnosis and treatment, while it still remains challenging due to the relatively low contrasts and ambiguous boundaries between abnormalities and normal regions. Additionally, the huge intra-class variances, alone with the high degree of visual similarities shared by inter-class abnormalities prevent the network from robust classification. To tackle these dilemmas, we propose an Adaptive Abnormal-aware Attention Network (Triple ANet) with Adaptive Dense Block (ADB) and Abnormal-aware Attention Module (AAM) for automatic WCE image analysis. ADB is designed to assign one attention score for each dense connection in dense blocks and to enhance useful features, while AAM aims to adaptively adjust the respective field according to the abnormal regions and help pay attention to abnormalities. Moreover, we propose a novel Angular Contrastive loss (AC Loss) to reduce the intra-class variances and enlarge the inter-class differences effectively. Our methods achieved 89.41% overall accuracy and showed better performance compared with state-of-the-art WCE image classification methods. The source code is available at https://github.com/Guo-Xiaoqing/Triple-ANet.

UR - https://www.scopus.com/pages/publications/85075629750

U2 - 10.1007/978-3-030-32239-7_33

DO - 10.1007/978-3-030-32239-7_33

M3 - Conference proceeding

AN - SCOPUS:85075629750

SN - 9783030322380

T3 - Lecture Notes in Computer Science

SP - 293

EP - 301

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2019

A2 - Shen, Dinggang

A2 - Yap, Pew-Thian

A2 - Liu, Tianming

A2 - Peters, Terry M.

A2 - Khan, Ali

A2 - Staib, Lawrence H.

A2 - Essert, Caroline

A2 - Zhou, Sean

PB - Springer Cham

Y2 - 13 October 2019 through 17 October 2019

ER -

Guo X, Yuan Y. Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification. In Shen D, Yap PT, Liu T, Peters TM, Khan A, Staib LH, Essert C, Zhou S, editors, Medical Image Computing and Computer Assisted Intervention – MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I. 1st ed. Springer Cham. 2019. p. 293-301. (Lecture Notes in Computer Science). doi: 10.1007/978-3-030-32239-7_33

Triple ANet: Adaptive Abnormal-aware Attention Network for WCE Image Classification

Abstract

Publication series

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this