Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers

Yangen Zhan; Hao Bian; Yang Chen; Xiu Li; Yongbing Zhang

doi:10.1109/ISBIC56247.2022.9854733

Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers

Yangen Zhan
, Hao Bian
, Yang Chen
, Xiu Li
, Yongbing Zhang^*

^*Corresponding author for this work

Tsinghua University
Harbin Institute of Technology Shenzhen

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

3 Scopus citations

Abstract

Artificial intelligence models have become increasingly promising in automated computed-aided cancer diagnos-tics. In this paper, a new deep learning method is proposed for solving breast tumor image classification in the BRIGHT Challenge. Given two types of data, regions of interest (ROIs) and whole slide images (WSIs), the proposed method first uti-lize ROI data to train a model that is able to select important image patches in WSI data and simultaneously capture a low-dimensional representation for image patches. With features extracted from important image patches in each WSI, another deep learning model following the Transformer framework is trained to perform the final WSI-level tumor classification, in which transfer learning is also employed to fully exploit ROI data. Evaluated on the test dataset, the proposed method achieves the 4th best results in the challenge. Ablation exper-iments are also carried out to analyze the proposed method in detail.

Original language	English
Title of host publication	ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665451727
DOIs	https://doi.org/10.1109/ISBIC56247.2022.9854733
State	Published - 2022
Externally published	Yes
Event	2022 IEEE International Symposium on Biomedical Imaging Challenges, ISBIC 2022 - Kolkata, India Duration: 28 Mar 2022 → 31 Mar 2022

Publication series

Name	ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings

Conference

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

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

BRIGHT Challenge
Multiple instance learning
Transformer
image classification
transfer learning

Access to Document

10.1109/ISBIC56247.2022.9854733

Cite this

Zhan, Y., Bian, H., Chen, Y., Li, X., & Zhang, Y. (2022). Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers. In ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings (ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISBIC56247.2022.9854733

Zhan, Yangen ; Bian, Hao ; Chen, Yang et al. / Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers. ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. (ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings).

@inproceedings{777c04432fd246dbb674347258642660,

title = "Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers",

abstract = "Artificial intelligence models have become increasingly promising in automated computed-aided cancer diagnos-tics. In this paper, a new deep learning method is proposed for solving breast tumor image classification in the BRIGHT Challenge. Given two types of data, regions of interest (ROIs) and whole slide images (WSIs), the proposed method first uti-lize ROI data to train a model that is able to select important image patches in WSI data and simultaneously capture a low-dimensional representation for image patches. With features extracted from important image patches in each WSI, another deep learning model following the Transformer framework is trained to perform the final WSI-level tumor classification, in which transfer learning is also employed to fully exploit ROI data. Evaluated on the test dataset, the proposed method achieves the 4th best results in the challenge. Ablation exper-iments are also carried out to analyze the proposed method in detail.",

keywords = "BRIGHT Challenge, Multiple instance learning, Transformer, image classification, transfer learning",

author = "Yangen Zhan and Hao Bian and Yang Chen and Xiu Li and Yongbing Zhang",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Symposium on Biomedical Imaging Challenges, ISBIC 2022 ; Conference date: 28-03-2022 Through 31-03-2022",

year = "2022",

doi = "10.1109/ISBIC56247.2022.9854733",

language = "英语",

series = "ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings",

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

booktitle = "ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings",

address = "美国",

}

Zhan, Y, Bian, H, Chen, Y, Li, X & Zhang, Y 2022, Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers. in ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings. ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings, Institute of Electrical and Electronics Engineers Inc., 2022 IEEE International Symposium on Biomedical Imaging Challenges, ISBIC 2022, Kolkata, India, 28/03/22. https://doi.org/10.1109/ISBIC56247.2022.9854733

Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers. / Zhan, Yangen; Bian, Hao; Chen, Yang et al.
ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. (ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings).

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

TY - GEN

T1 - Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers

AU - Zhan, Yangen

AU - Bian, Hao

AU - Chen, Yang

AU - Li, Xiu

AU - Zhang, Yongbing

PY - 2022

Y1 - 2022

N2 - Artificial intelligence models have become increasingly promising in automated computed-aided cancer diagnos-tics. In this paper, a new deep learning method is proposed for solving breast tumor image classification in the BRIGHT Challenge. Given two types of data, regions of interest (ROIs) and whole slide images (WSIs), the proposed method first uti-lize ROI data to train a model that is able to select important image patches in WSI data and simultaneously capture a low-dimensional representation for image patches. With features extracted from important image patches in each WSI, another deep learning model following the Transformer framework is trained to perform the final WSI-level tumor classification, in which transfer learning is also employed to fully exploit ROI data. Evaluated on the test dataset, the proposed method achieves the 4th best results in the challenge. Ablation exper-iments are also carried out to analyze the proposed method in detail.

AB - Artificial intelligence models have become increasingly promising in automated computed-aided cancer diagnos-tics. In this paper, a new deep learning method is proposed for solving breast tumor image classification in the BRIGHT Challenge. Given two types of data, regions of interest (ROIs) and whole slide images (WSIs), the proposed method first uti-lize ROI data to train a model that is able to select important image patches in WSI data and simultaneously capture a low-dimensional representation for image patches. With features extracted from important image patches in each WSI, another deep learning model following the Transformer framework is trained to perform the final WSI-level tumor classification, in which transfer learning is also employed to fully exploit ROI data. Evaluated on the test dataset, the proposed method achieves the 4th best results in the challenge. Ablation exper-iments are also carried out to analyze the proposed method in detail.

KW - BRIGHT Challenge

KW - Multiple instance learning

KW - Transformer

KW - image classification

KW - transfer learning

UR - https://www.scopus.com/pages/publications/85137273145

U2 - 10.1109/ISBIC56247.2022.9854733

DO - 10.1109/ISBIC56247.2022.9854733

M3 - 会议稿件

AN - SCOPUS:85137273145

T3 - ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings

BT - ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 IEEE International Symposium on Biomedical Imaging Challenges, ISBIC 2022

Y2 - 28 March 2022 through 31 March 2022

ER -

Zhan Y, Bian H, Chen Y, Li X, Zhang Y. Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers. In ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2022. (ISBIC 2022 - International Symposium on Biomedical Imaging Challenges, Proceedings). doi: 10.1109/ISBIC56247.2022.9854733

Breast Tumor Image Classification in Bright Challenge VIA Multiple Instance Learning and Deep Transformers

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this