Breast tissue segmentation and mammographic risk scoring using deep learning

Peter Kersten Petersen; Mads Nielsen; Pengfei Diao; Nico Karssemeijer; Martin Lillholm

doi:10.1007/978-3-319-07887-8_13

Breast tissue segmentation and mammographic risk scoring using deep learning

Peter Kersten Petersen, Mads Nielsen, Pengfei Diao, Nico Karssemeijer, Martin Lillholm

Department of Computer Science

34 Citations (Scopus)

Abstract

Mammographic scoring of density and texture are established methods to relate to the risk of breast cancer. We present a method that learns descriptive features from unlabeled mammograms and, using these learned features as the input to a simple classifier, address the following tasks: i) breast tissue segmentation ii) scoring of percentage mammographic density (PMD), and iii) scoring of mammographic texture (MT). Our results suggest that the learned PMD scores correlate well to manual ones, and that the learned MT scores are more related to future cancer risk than both manual and automatic PMD scores.

Original language	English
Title of host publication	Breast imaging : 12th International Workshop, IWDM 2014, Gifu City, Japan, June 29 – July 2, 2014. Proceedings
Editors	Hiroshi Fujita, Takeshi Hara, Chisako Muramatsu
Number of pages	7
Publisher	Springer Science+Business Media
Publication date	2014
Pages	88-94
ISBN (Print)	978-3-319-07886-1
ISBN (Electronic)	978-3-319-07887-8
DOIs	https://doi.org/10.1007/978-3-319-07887-8_13
Publication status	Published - 2014
Event	International Workshop, IWDM 2014 - Gifu City, Japan Duration: 29 Jun 2014 → 2 Jul 2014 Conference number: 12

Conference

Conference	International Workshop, IWDM 2014
Number	12
Country/Territory	Japan
City	Gifu City
Period	29/06/2014 → 02/07/2014

Series	Lecture notes in computer science
Volume	8539
ISSN	0302-9743

UN SDGs

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

Access to Document

10.1007/978-3-319-07887-8_13

Cite this

Petersen, P. K., Nielsen, M., Diao, P., Karssemeijer, N., & Lillholm, M. (2014). Breast tissue segmentation and mammographic risk scoring using deep learning. In H. Fujita, T. Hara, & C. Muramatsu (Eds.), Breast imaging: 12th International Workshop, IWDM 2014, Gifu City, Japan, June 29 – July 2, 2014. Proceedings (pp. 88-94). Springer Science+Business Media. Lecture notes in computer science Vol. 8539 https://doi.org/10.1007/978-3-319-07887-8_13

Breast tissue segmentation and mammographic risk scoring using deep learning. / Petersen, Peter Kersten; Nielsen, Mads ; Diao, Pengfei et al.

Breast imaging: 12th International Workshop, IWDM 2014, Gifu City, Japan, June 29 – July 2, 2014. Proceedings. ed. / Hiroshi Fujita; Takeshi Hara; Chisako Muramatsu. Springer Science+Business Media, 2014. p. 88-94 (Lecture notes in computer science, Vol. 8539).

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

Petersen, PK, Nielsen, M , Diao, P, Karssemeijer, N & Lillholm, M 2014, Breast tissue segmentation and mammographic risk scoring using deep learning. in H Fujita, T Hara & C Muramatsu (eds), Breast imaging: 12th International Workshop, IWDM 2014, Gifu City, Japan, June 29 – July 2, 2014. Proceedings. Springer Science+Business Media, Lecture notes in computer science, vol. 8539, pp. 88-94, International Workshop, IWDM 2014, Gifu City, Japan, 29/06/2014. https://doi.org/10.1007/978-3-319-07887-8_13

Petersen PK, Nielsen M , Diao P, Karssemeijer N, Lillholm M. Breast tissue segmentation and mammographic risk scoring using deep learning. In Fujita H, Hara T, Muramatsu C, editors, Breast imaging: 12th International Workshop, IWDM 2014, Gifu City, Japan, June 29 – July 2, 2014. Proceedings. Springer Science+Business Media. 2014. p. 88-94. (Lecture notes in computer science, Vol. 8539). doi: 10.1007/978-3-319-07887-8_13

Petersen, Peter Kersten ; Nielsen, Mads ; Diao, Pengfei et al. / Breast tissue segmentation and mammographic risk scoring using deep learning. Breast imaging: 12th International Workshop, IWDM 2014, Gifu City, Japan, June 29 – July 2, 2014. Proceedings. editor / Hiroshi Fujita ; Takeshi Hara ; Chisako Muramatsu. Springer Science+Business Media, 2014. pp. 88-94 (Lecture notes in computer science, Vol. 8539).

@inproceedings{938dbc4aab944e6eba3c3781255ed831,

title = "Breast tissue segmentation and mammographic risk scoring using deep learning",

abstract = "Mammographic scoring of density and texture are established methods to relate to the risk of breast cancer. We present a method that learns descriptive features from unlabeled mammograms and, using these learned features as the input to a simple classifier, address the following tasks: i) breast tissue segmentation ii) scoring of percentage mammographic density (PMD), and iii) scoring of mammographic texture (MT). Our results suggest that the learned PMD scores correlate well to manual ones, and that the learned MT scores are more related to future cancer risk than both manual and automatic PMD scores.",

keywords = "Unsupervised feature learning, deep learning, breast cancer, mammograms, prognosis, risk factor, segmentation",

author = "Petersen, {Peter Kersten} and Mads Nielsen and Pengfei Diao and Nico Karssemeijer and Martin Lillholm",

year = "2014",

doi = "10.1007/978-3-319-07887-8_13",

language = "English",

isbn = "978-3-319-07886-1",

series = "Lecture notes in computer science",

publisher = "Springer Science+Business Media",

pages = "88--94",

editor = "Hiroshi Fujita and Takeshi Hara and Chisako Muramatsu",

booktitle = "Breast imaging",

address = "Singapore",

note = "International Workshop, IWDM 2014 ; Conference date: 29-06-2014 Through 02-07-2014",

}

TY - GEN

T1 - Breast tissue segmentation and mammographic risk scoring using deep learning

AU - Petersen, Peter Kersten

AU - Nielsen, Mads

AU - Diao, Pengfei

AU - Karssemeijer, Nico

AU - Lillholm, Martin

N1 - Conference code: 12

PY - 2014

Y1 - 2014

N2 - Mammographic scoring of density and texture are established methods to relate to the risk of breast cancer. We present a method that learns descriptive features from unlabeled mammograms and, using these learned features as the input to a simple classifier, address the following tasks: i) breast tissue segmentation ii) scoring of percentage mammographic density (PMD), and iii) scoring of mammographic texture (MT). Our results suggest that the learned PMD scores correlate well to manual ones, and that the learned MT scores are more related to future cancer risk than both manual and automatic PMD scores.

AB - Mammographic scoring of density and texture are established methods to relate to the risk of breast cancer. We present a method that learns descriptive features from unlabeled mammograms and, using these learned features as the input to a simple classifier, address the following tasks: i) breast tissue segmentation ii) scoring of percentage mammographic density (PMD), and iii) scoring of mammographic texture (MT). Our results suggest that the learned PMD scores correlate well to manual ones, and that the learned MT scores are more related to future cancer risk than both manual and automatic PMD scores.

KW - Unsupervised feature learning

KW - deep learning

KW - breast cancer

KW - mammograms

KW - prognosis

KW - risk factor

KW - segmentation

U2 - 10.1007/978-3-319-07887-8_13

DO - 10.1007/978-3-319-07887-8_13

M3 - Article in proceedings

SN - 978-3-319-07886-1

T3 - Lecture notes in computer science

SP - 88

EP - 94

BT - Breast imaging

A2 - Fujita, Hiroshi

A2 - Hara, Takeshi

A2 - Muramatsu, Chisako

PB - Springer Science+Business Media

T2 - International Workshop, IWDM 2014

Y2 - 29 June 2014 through 2 July 2014

ER -

Breast tissue segmentation and mammographic risk scoring using deep learning

Abstract

Conference

UN SDGs

Access to Document

Fingerprint

Cite this