Imbalanced histopathology image classification using deep feature graph attention network

Cong Cong; Yixing Yang; Sidong Liu; Maurice Pagnucco; Yang Song

doi:10.1109/ISBI52829.2022.9761619

Imbalanced histopathology image classification using deep feature graph attention network

Cong Cong, Yixing Yang, Sidong Liu, Maurice Pagnucco, Yang Song

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

4 Citations (Scopus)

Abstract

Recent deep learning studies have shown great progress in metastasis detection over histopathology whole slide images (WSIs). As WSIs are extremely large, most of existing studies adopt patch level analysis, leveraging spatial context to enhance patch-wise classification. However, class imbalance in patch distribution may result in an exceedingly large number of false-negatives, thereby worsening the WSI level classification performance. In this paper, we propose a novel framework for classification in class imbalanced datasets, which adopts a graph attention network to capture feature dependent interactions and a minority preferred inference mechanism for patch-level classification. Our experiments on CAMELYON16 show that the proposed method substantially improves detection of the minority class (tumour) under a highly imbalanced class distribution.

Original language	English
Title of host publication	IEEE ISBI 2022 Proceedings
Subtitle of host publication	2022 IEEE International Symposium on Biomedical Imaging
Place of Publication	Kolkata
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	1-4
Number of pages	4
ISBN (Electronic)	9781665429238
ISBN (Print)	9781665429245
DOIs	https://doi.org/10.1109/ISBI52829.2022.9761619
Publication status	Published - 28 Mar 2022
Event	19th IEEE International Symposium on Biomedical Imaging, ISBI 2022 - Kolkata, India Duration: 28 Mar 2022 → 31 Mar 2022

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
Volume	2022-March
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	19th IEEE International Symposium on Biomedical Imaging, ISBI 2022
Country/Territory	India
City	Kolkata
Period	28/03/22 → 31/03/22

Keywords

Graph Attention Networks
Intra-Class Feature Enhancement
Metastasis Detection

Access to Document

10.1109/ISBI52829.2022.9761619Licence: Unspecified

Cite this

Cong, C., Yang, Y., Liu, S., Pagnucco, M., & Song, Y. (2022). Imbalanced histopathology image classification using deep feature graph attention network. In IEEE ISBI 2022 Proceedings: 2022 IEEE International Symposium on Biomedical Imaging (pp. 1-4). (Proceedings - International Symposium on Biomedical Imaging; Vol. 2022-March). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ISBI52829.2022.9761619

Cong, Cong ; Yang, Yixing ; Liu, Sidong et al. / Imbalanced histopathology image classification using deep feature graph attention network. IEEE ISBI 2022 Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. Kolkata : Institute of Electrical and Electronics Engineers (IEEE), 2022. pp. 1-4 (Proceedings - International Symposium on Biomedical Imaging).

@inproceedings{039d1d25ccd2428db8efea3a25ccd01c,

title = "Imbalanced histopathology image classification using deep feature graph attention network",

abstract = "Recent deep learning studies have shown great progress in metastasis detection over histopathology whole slide images (WSIs). As WSIs are extremely large, most of existing studies adopt patch level analysis, leveraging spatial context to enhance patch-wise classification. However, class imbalance in patch distribution may result in an exceedingly large number of false-negatives, thereby worsening the WSI level classification performance. In this paper, we propose a novel framework for classification in class imbalanced datasets, which adopts a graph attention network to capture feature dependent interactions and a minority preferred inference mechanism for patch-level classification. Our experiments on CAMELYON16 show that the proposed method substantially improves detection of the minority class (tumour) under a highly imbalanced class distribution.",

keywords = "Graph Attention Networks, Intra-Class Feature Enhancement, Metastasis Detection",

author = "Cong Cong and Yixing Yang and Sidong Liu and Maurice Pagnucco and Yang Song",

year = "2022",

month = mar,

day = "28",

doi = "10.1109/ISBI52829.2022.9761619",

language = "English",

isbn = "9781665429245",

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

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "1--4",

booktitle = "IEEE ISBI 2022 Proceedings",

address = "United States",

note = "19th IEEE International Symposium on Biomedical Imaging, ISBI 2022 ; Conference date: 28-03-2022 Through 31-03-2022",

}

Cong, C, Yang, Y, Liu, S, Pagnucco, M & Song, Y 2022, Imbalanced histopathology image classification using deep feature graph attention network. in IEEE ISBI 2022 Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. Proceedings - International Symposium on Biomedical Imaging, vol. 2022-March, Institute of Electrical and Electronics Engineers (IEEE), Kolkata, pp. 1-4, 19th IEEE International Symposium on Biomedical Imaging, ISBI 2022, Kolkata, India, 28/03/22. https://doi.org/10.1109/ISBI52829.2022.9761619

Imbalanced histopathology image classification using deep feature graph attention network. / Cong, Cong; Yang, Yixing; Liu, Sidong et al.
IEEE ISBI 2022 Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. Kolkata: Institute of Electrical and Electronics Engineers (IEEE), 2022. p. 1-4 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2022-March).

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

TY - GEN

T1 - Imbalanced histopathology image classification using deep feature graph attention network

AU - Cong, Cong

AU - Yang, Yixing

AU - Liu, Sidong

AU - Pagnucco, Maurice

AU - Song, Yang

PY - 2022/3/28

Y1 - 2022/3/28

N2 - Recent deep learning studies have shown great progress in metastasis detection over histopathology whole slide images (WSIs). As WSIs are extremely large, most of existing studies adopt patch level analysis, leveraging spatial context to enhance patch-wise classification. However, class imbalance in patch distribution may result in an exceedingly large number of false-negatives, thereby worsening the WSI level classification performance. In this paper, we propose a novel framework for classification in class imbalanced datasets, which adopts a graph attention network to capture feature dependent interactions and a minority preferred inference mechanism for patch-level classification. Our experiments on CAMELYON16 show that the proposed method substantially improves detection of the minority class (tumour) under a highly imbalanced class distribution.

AB - Recent deep learning studies have shown great progress in metastasis detection over histopathology whole slide images (WSIs). As WSIs are extremely large, most of existing studies adopt patch level analysis, leveraging spatial context to enhance patch-wise classification. However, class imbalance in patch distribution may result in an exceedingly large number of false-negatives, thereby worsening the WSI level classification performance. In this paper, we propose a novel framework for classification in class imbalanced datasets, which adopts a graph attention network to capture feature dependent interactions and a minority preferred inference mechanism for patch-level classification. Our experiments on CAMELYON16 show that the proposed method substantially improves detection of the minority class (tumour) under a highly imbalanced class distribution.

KW - Graph Attention Networks

KW - Intra-Class Feature Enhancement

KW - Metastasis Detection

UR - http://www.scopus.com/inward/record.url?scp=85129631400&partnerID=8YFLogxK

U2 - 10.1109/ISBI52829.2022.9761619

DO - 10.1109/ISBI52829.2022.9761619

M3 - Conference proceeding contribution

AN - SCOPUS:85129631400

SN - 9781665429245

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 1

EP - 4

BT - IEEE ISBI 2022 Proceedings

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - Kolkata

T2 - 19th IEEE International Symposium on Biomedical Imaging, ISBI 2022

Y2 - 28 March 2022 through 31 March 2022

ER -

Cong C, Yang Y, Liu S, Pagnucco M, Song Y. Imbalanced histopathology image classification using deep feature graph attention network. In IEEE ISBI 2022 Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. Kolkata: Institute of Electrical and Electronics Engineers (IEEE). 2022. p. 1-4. (Proceedings - International Symposium on Biomedical Imaging). doi: 10.1109/ISBI52829.2022.9761619

Imbalanced histopathology image classification using deep feature graph attention network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

AI-Assisted Digital Histopathology Image Computing for Tumor Diagnosis

Cite this

Imbalanced histopathology image classification using deep feature graph attention network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Projects

AI-Assisted Digital Histopathology Image Computing for Tumor Diagnosis

Cite this