Globally-Aware Multiple Instance Classifier for Breast Cancer Screening

Yiqiu Shen; Nan Wu; Jason Phang; Jungkyu Park; Gene Kim; Linda Moy; Kyunghyun Cho; Krzysztof J. Geras

doi:10.1007/978-3-030-32692-0_3

Globally-Aware Multiple Instance Classifier for Breast Cancer Screening

Yiqiu Shen, Nan Wu, Jason Phang, Jungkyu Park, Gene Kim, Linda Moy, Kyunghyun Cho, Krzysztof J. Geras

Computer Science

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

Abstract

Deep learning models designed for visual classification tasks on natural images have become prevalent in medical image analysis. However, medical images differ from typical natural images in many ways, such as significantly higher resolutions and smaller regions of interest. Moreover, both the global structure and local details play important roles in medical image analysis tasks. To address these unique properties of medical images, we propose a neural network that is able to classify breast cancer lesions utilizing information from both a global saliency map and multiple local patches. The proposed model outperforms the ResNet-based baseline and achieves radiologist-level performance in the interpretation of screening mammography. Although our model is trained only with image-level labels, it is able to generate pixel-level saliency maps that provide localization of possible malignant findings.

Original language	English (US)
Title of host publication	Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings
Editors	Heung-Il Suk, Mingxia Liu, Chunfeng Lian, Pingkun Yan
Publisher	Springer
Pages	18-26
Number of pages	9
Volume	11861
ISBN (Print)	9783030326913
DOIs	https://doi.org/10.1007/978-3-030-32692-0_3
State	Published - 2019
Event	10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China Duration: Oct 13 2019 → Oct 13 2019

Publication series

Name	Machine learning in medical imaging. MLMI (Workshop)

Conference

Conference	10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Country/Territory	China
City	Shenzhen
Period	10/13/19 → 10/13/19

Keywords

Breast cancer screening
Deep learning
High-resolution image classification
Neural networks
Weakly supervised localization

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-030-32692-0_3

Cite this

Shen, Y., Wu, N., Phang, J., Park, J., Kim, G., Moy, L., Cho, K., & Geras, K. J. (2019). Globally-Aware Multiple Instance Classifier for Breast Cancer Screening. In H.-I. Suk, M. Liu, C. Lian, & P. Yan (Eds.), Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings (Vol. 11861, pp. 18-26). (Machine learning in medical imaging. MLMI (Workshop)). Springer. https://doi.org/10.1007/978-3-030-32692-0_3

Globally-Aware Multiple Instance Classifier for Breast Cancer Screening. / Shen, Yiqiu; Wu, Nan; Phang, Jason et al.
Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. ed. / Heung-Il Suk; Mingxia Liu; Chunfeng Lian; Pingkun Yan. Vol. 11861 Springer, 2019. p. 18-26 (Machine learning in medical imaging. MLMI (Workshop)).

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

Shen, Y, Wu, N, Phang, J, Park, J, Kim, G, Moy, L, Cho, K & Geras, KJ 2019, Globally-Aware Multiple Instance Classifier for Breast Cancer Screening. in H-I Suk, M Liu, C Lian & P Yan (eds), Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. vol. 11861, Machine learning in medical imaging. MLMI (Workshop), Springer, pp. 18-26, 10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019, Shenzhen, China, 10/13/19. https://doi.org/10.1007/978-3-030-32692-0_3

Shen Y, Wu N, Phang J, Park J, Kim G, Moy L et al. Globally-Aware Multiple Instance Classifier for Breast Cancer Screening. In Suk HI, Liu M, Lian C, Yan P, editors, Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. Vol. 11861. Springer. 2019. p. 18-26. (Machine learning in medical imaging. MLMI (Workshop)). doi: 10.1007/978-3-030-32692-0_3

Shen, Yiqiu ; Wu, Nan ; Phang, Jason et al. / Globally-Aware Multiple Instance Classifier for Breast Cancer Screening. Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings. editor / Heung-Il Suk ; Mingxia Liu ; Chunfeng Lian ; Pingkun Yan. Vol. 11861 Springer, 2019. pp. 18-26 (Machine learning in medical imaging. MLMI (Workshop)).

@inproceedings{566d54ce46294528b139f97d79525ecc,

title = "Globally-Aware Multiple Instance Classifier for Breast Cancer Screening",

abstract = "Deep learning models designed for visual classification tasks on natural images have become prevalent in medical image analysis. However, medical images differ from typical natural images in many ways, such as significantly higher resolutions and smaller regions of interest. Moreover, both the global structure and local details play important roles in medical image analysis tasks. To address these unique properties of medical images, we propose a neural network that is able to classify breast cancer lesions utilizing information from both a global saliency map and multiple local patches. The proposed model outperforms the ResNet-based baseline and achieves radiologist-level performance in the interpretation of screening mammography. Although our model is trained only with image-level labels, it is able to generate pixel-level saliency maps that provide localization of possible malignant findings.",

keywords = "Breast cancer screening, Deep learning, High-resolution image classification, Neural networks, Weakly supervised localization",

author = "Yiqiu Shen and Nan Wu and Jason Phang and Jungkyu Park and Gene Kim and Linda Moy and Kyunghyun Cho and Geras, {Krzysztof J.}",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Nature Switzerland AG.; 10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 ; Conference date: 13-10-2019 Through 13-10-2019",

year = "2019",

doi = "10.1007/978-3-030-32692-0_3",

language = "English (US)",

isbn = "9783030326913",

volume = "11861",

series = "Machine learning in medical imaging. MLMI (Workshop)",

publisher = "Springer",

pages = "18--26",

editor = "Heung-Il Suk and Mingxia Liu and Chunfeng Lian and Pingkun Yan",

booktitle = "Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings",

}

TY - GEN

T1 - Globally-Aware Multiple Instance Classifier for Breast Cancer Screening

AU - Shen, Yiqiu

AU - Wu, Nan

AU - Phang, Jason

AU - Park, Jungkyu

AU - Kim, Gene

AU - Moy, Linda

AU - Cho, Kyunghyun

AU - Geras, Krzysztof J.

PY - 2019

Y1 - 2019

N2 - Deep learning models designed for visual classification tasks on natural images have become prevalent in medical image analysis. However, medical images differ from typical natural images in many ways, such as significantly higher resolutions and smaller regions of interest. Moreover, both the global structure and local details play important roles in medical image analysis tasks. To address these unique properties of medical images, we propose a neural network that is able to classify breast cancer lesions utilizing information from both a global saliency map and multiple local patches. The proposed model outperforms the ResNet-based baseline and achieves radiologist-level performance in the interpretation of screening mammography. Although our model is trained only with image-level labels, it is able to generate pixel-level saliency maps that provide localization of possible malignant findings.

AB - Deep learning models designed for visual classification tasks on natural images have become prevalent in medical image analysis. However, medical images differ from typical natural images in many ways, such as significantly higher resolutions and smaller regions of interest. Moreover, both the global structure and local details play important roles in medical image analysis tasks. To address these unique properties of medical images, we propose a neural network that is able to classify breast cancer lesions utilizing information from both a global saliency map and multiple local patches. The proposed model outperforms the ResNet-based baseline and achieves radiologist-level performance in the interpretation of screening mammography. Although our model is trained only with image-level labels, it is able to generate pixel-level saliency maps that provide localization of possible malignant findings.

KW - Breast cancer screening

KW - Deep learning

KW - High-resolution image classification

KW - Neural networks

KW - Weakly supervised localization

UR - http://www.scopus.com/inward/record.url?scp=85075700901&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075700901&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-32692-0_3

DO - 10.1007/978-3-030-32692-0_3

M3 - Conference contribution

C2 - 32149282

AN - SCOPUS:85075700901

SN - 9783030326913

VL - 11861

T3 - Machine learning in medical imaging. MLMI (Workshop)

SP - 18

EP - 26

BT - Machine Learning in Medical Imaging - 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Proceedings

A2 - Suk, Heung-Il

A2 - Liu, Mingxia

A2 - Lian, Chunfeng

A2 - Yan, Pingkun

PB - Springer

T2 - 10th International Workshop on Machine Learning in Medical Imaging, MLMI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019

Y2 - 13 October 2019 through 13 October 2019

ER -

Globally-Aware Multiple Instance Classifier for Breast Cancer Screening

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this