Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images

Muhammad Waqas; Tomas Maul; Iman Yi Liao; Amr Ahmed

doi:10.1109/CISP-BMEI56279.2022.9980033

Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images

Muhammad Waqas^*
, Tomas Maul
, Iman Yi Liao
, Amr Ahmed

^*Corresponding author for this work

University of Nottingham Malaysia
University of Malakand
University of Nottingham

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding (ISBN) › peer-review

6 Citations (Scopus)

Abstract

Out of all cancers among female patients, breast cancer is on top in terms of prevalence. Many deep learning approaches have been investigated for the classification of histopathological images of breast cancer, however most tend to be either too complex or too large to adopt at a clinical level. In this study we propose a lightweight network, based on knowledge distillation, which performs almost the same as the teacher network but with significantly fewer parameters. Our student network consists of inverted residual blocks of MobileNetV2 and the ghost module. We performed our experiments on BreakHis, BACH and Kaggle breast cancer histopathological imaging datasets. The results show that different versions of our proposed lightweight student architecture perform with similar accuracy levels compared with the teacher network, while using significantly fewer parameters.

Original language	English
Title of host publication	Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022
Editors	Xin Chen, Lin Cao, Qingli Li, Yan Wang, Lipo Wang
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665488877
ISBN (Print)	9781665488877
DOIs	https://doi.org/10.1109/CISP-BMEI56279.2022.9980033
Publication status	Published - 21 Dec 2022
Event	15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022 - Beijing, China Duration: 5 Nov 2022 → 7 Nov 2022

Publication series

Name	Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022

Conference

Conference	15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022
Country/Territory	China
City	Beijing
Period	5/11/22 → 7/11/22

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

Breast cancer
Histopathological Images
Knowledge Distillation

Access to Document

10.1109/CISP-BMEI56279.2022.9980033

Cite this

Waqas, M., Maul, T., Liao, I. Y., & Ahmed, A. (2022). Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images. In X. Chen, L. Cao, Q. Li, Y. Wang, & L. Wang (Eds.), Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022 (Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CISP-BMEI56279.2022.9980033

Waqas, Muhammad ; Maul, Tomas ; Liao, Iman Yi et al. / Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images. Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022. editor / Xin Chen ; Lin Cao ; Qingli Li ; Yan Wang ; Lipo Wang. Institute of Electrical and Electronics Engineers Inc., 2022. (Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022).

@inproceedings{d1008722a0c849539ccbbb6372a6a8d6,

title = "Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images",

abstract = "Out of all cancers among female patients, breast cancer is on top in terms of prevalence. Many deep learning approaches have been investigated for the classification of histopathological images of breast cancer, however most tend to be either too complex or too large to adopt at a clinical level. In this study we propose a lightweight network, based on knowledge distillation, which performs almost the same as the teacher network but with significantly fewer parameters. Our student network consists of inverted residual blocks of MobileNetV2 and the ghost module. We performed our experiments on BreakHis, BACH and Kaggle breast cancer histopathological imaging datasets. The results show that different versions of our proposed lightweight student architecture perform with similar accuracy levels compared with the teacher network, while using significantly fewer parameters.",

keywords = "Breast cancer, Histopathological Images, Knowledge Distillation",

author = "Muhammad Waqas and Tomas Maul and Liao, \{Iman Yi\} and Amr Ahmed",

note = "Funding Information: This work would not have been possible without the financial support of the University of Malakand, Higher Education Commission (Government of Pakistan) and the University of Nottingham Malaysia. Publisher Copyright: {\textcopyright} 2022 IEEE.; 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022 ; Conference date: 05-11-2022 Through 07-11-2022",

year = "2022",

month = dec,

day = "21",

doi = "10.1109/CISP-BMEI56279.2022.9980033",

language = "English",

isbn = "9781665488877",

series = "Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022",

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

editor = "Xin Chen and Lin Cao and Qingli Li and Yan Wang and Lipo Wang",

booktitle = "Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022",

address = "United States",

}

Waqas, M, Maul, T, Liao, IY & Ahmed, A 2022, Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images. in X Chen, L Cao, Q Li, Y Wang & L Wang (eds), Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022. Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022, Institute of Electrical and Electronics Engineers Inc., 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022, Beijing, China, 5/11/22. https://doi.org/10.1109/CISP-BMEI56279.2022.9980033

Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images. / Waqas, Muhammad; Maul, Tomas; Liao, Iman Yi et al.
Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022. ed. / Xin Chen; Lin Cao; Qingli Li; Yan Wang; Lipo Wang. Institute of Electrical and Electronics Engineers Inc., 2022. (Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding (ISBN) › peer-review

TY - GEN

T1 - Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images

AU - Waqas, Muhammad

AU - Maul, Tomas

AU - Liao, Iman Yi

AU - Ahmed, Amr

N1 - Funding Information: This work would not have been possible without the financial support of the University of Malakand, Higher Education Commission (Government of Pakistan) and the University of Nottingham Malaysia. Publisher Copyright: © 2022 IEEE.

PY - 2022/12/21

Y1 - 2022/12/21

N2 - Out of all cancers among female patients, breast cancer is on top in terms of prevalence. Many deep learning approaches have been investigated for the classification of histopathological images of breast cancer, however most tend to be either too complex or too large to adopt at a clinical level. In this study we propose a lightweight network, based on knowledge distillation, which performs almost the same as the teacher network but with significantly fewer parameters. Our student network consists of inverted residual blocks of MobileNetV2 and the ghost module. We performed our experiments on BreakHis, BACH and Kaggle breast cancer histopathological imaging datasets. The results show that different versions of our proposed lightweight student architecture perform with similar accuracy levels compared with the teacher network, while using significantly fewer parameters.

AB - Out of all cancers among female patients, breast cancer is on top in terms of prevalence. Many deep learning approaches have been investigated for the classification of histopathological images of breast cancer, however most tend to be either too complex or too large to adopt at a clinical level. In this study we propose a lightweight network, based on knowledge distillation, which performs almost the same as the teacher network but with significantly fewer parameters. Our student network consists of inverted residual blocks of MobileNetV2 and the ghost module. We performed our experiments on BreakHis, BACH and Kaggle breast cancer histopathological imaging datasets. The results show that different versions of our proposed lightweight student architecture perform with similar accuracy levels compared with the teacher network, while using significantly fewer parameters.

KW - Breast cancer

KW - Histopathological Images

KW - Knowledge Distillation

UR - https://www.scopus.com/pages/publications/85146216847

UR - https://www.scopus.com/pages/publications/85146216847#tab=citedBy

U2 - 10.1109/CISP-BMEI56279.2022.9980033

DO - 10.1109/CISP-BMEI56279.2022.9980033

M3 - Conference proceeding (ISBN)

AN - SCOPUS:85146216847

SN - 9781665488877

T3 - Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022

BT - Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022

A2 - Chen, Xin

A2 - Cao, Lin

A2 - Li, Qingli

A2 - Wang, Yan

A2 - Wang, Lipo

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022

Y2 - 5 November 2022 through 7 November 2022

ER -

Waqas M, Maul T, Liao IY, Ahmed A. Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images. In Chen X, Cao L, Li Q, Wang Y, Wang L, editors, Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022. Institute of Electrical and Electronics Engineers Inc. 2022. (Proceedings - 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2022). doi: 10.1109/CISP-BMEI56279.2022.9980033

Lightweight Deep Network for the Classification of Breast Cancer Histopathological Images

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this