Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation

Gopal Karemore; Brad M. Keller; Huen Oh; Julia Tchou; Mads Nielsen; Emily F. Conant; Despina Kontos

doi:10.1007/978-3-642-31271-7_77

Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation

Gopal Karemore, Brad M. Keller, Huen Oh, Julia Tchou, Mads Nielsen, Emily F. Conant, Despina Kontos

Department of Computer Science

4 Citations (Scopus)

Abstract

We investigate the potential of mammographic parenchymal texture as a surrogate marker of the risk to develop Estrogen Receptor (ER) sub-type specific breast cancer. A case-control study was performed, including 118 cancer cases stratified by ER receptor status and 354 age-matched controls. Digital mammographic (DM) images were retrospectively collected and analyzed under HIPAA and IRB approval. The performance of the texture features was compared to that of the standard mammographic density measures. We observed that breast percent density PD% and parenchymal texture features can both distinguish between cancer cases and controls (A _z > 0.70). However, for ER subtype-specific classification, PD% alone does not provide sufficient classification (A _z = 0.60), while texture features have significant classification performance (A _z = 0.70). Combining breast density with texture features achieves the best performance (A _z = 0.71). These findings suggest that mammographic texture analysis may have value for sub-type specific breast cancer risk assessment.

Original language	English
Title of host publication	Breast Imaging : 11th International Workshop, IWDM 2012, Philadelphia, PA, USA, July 8-11, 2012. Proceedings
Editors	Andrew D. A. Maidment, Predrag R. Bakic, Sara Gavenonis
Number of pages	8
Publisher	Springer
Publication date	2012
Pages	596-603
ISBN (Print)	978-3-642-31270-0
ISBN (Electronic)	978-3-642-31271-7
DOIs	https://doi.org/10.1007/978-3-642-31271-7_77
Publication status	Published - 2012
Event	11th International Workshop on Breast Imaging - Philadelphia, PA, United States Duration: 8 Jul 2012 → 11 Jul 2012 Conference number: 11

Workshop

Workshop	11th International Workshop on Breast Imaging
Number	11
Country/Territory	United States
City	Philadelphia, PA
Period	08/07/2012 → 11/07/2012

Series	Lecture notes in computer science
Volume	7361
ISSN	0302-9743

UN SDGs

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

Access to Document

10.1007/978-3-642-31271-7_77

Cite this

Karemore, G., Keller, B. M., Oh, H., Tchou, J., Nielsen, M., Conant, E. F., & Kontos, D. (2012). Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation. In A. D. A. Maidment, P. R. Bakic, & S. Gavenonis (Eds.), Breast Imaging: 11th International Workshop, IWDM 2012, Philadelphia, PA, USA, July 8-11, 2012. Proceedings (pp. 596-603). Springer. Lecture notes in computer science Vol. 7361 https://doi.org/10.1007/978-3-642-31271-7_77

Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation. / Karemore, Gopal; Keller, Brad M.; Oh, Huen et al.

Breast Imaging: 11th International Workshop, IWDM 2012, Philadelphia, PA, USA, July 8-11, 2012. Proceedings. ed. / Andrew D. A. Maidment; Predrag R. Bakic; Sara Gavenonis. Springer, 2012. p. 596-603 (Lecture notes in computer science, Vol. 7361).

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

Karemore, G, Keller, BM, Oh, H, Tchou, J, Nielsen, M, Conant, EF & Kontos, D 2012, Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation. in ADA Maidment, PR Bakic & S Gavenonis (eds), Breast Imaging: 11th International Workshop, IWDM 2012, Philadelphia, PA, USA, July 8-11, 2012. Proceedings. Springer, Lecture notes in computer science, vol. 7361, pp. 596-603, 11th International Workshop on Breast Imaging, Philadelphia, PA, United States, 08/07/2012. https://doi.org/10.1007/978-3-642-31271-7_77

Karemore G, Keller BM, Oh H, Tchou J, Nielsen M, Conant EF et al. Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation. In Maidment ADA, Bakic PR, Gavenonis S, editors, Breast Imaging: 11th International Workshop, IWDM 2012, Philadelphia, PA, USA, July 8-11, 2012. Proceedings. Springer. 2012. p. 596-603. (Lecture notes in computer science, Vol. 7361). doi: 10.1007/978-3-642-31271-7_77

Karemore, Gopal ; Keller, Brad M. ; Oh, Huen et al. / Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation. Breast Imaging: 11th International Workshop, IWDM 2012, Philadelphia, PA, USA, July 8-11, 2012. Proceedings. editor / Andrew D. A. Maidment ; Predrag R. Bakic ; Sara Gavenonis. Springer, 2012. pp. 596-603 (Lecture notes in computer science, Vol. 7361).

@inproceedings{7b1a89acae41460396bfbe880efb3587,

title = "Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation",

abstract = "We investigate the potential of mammographic parenchymal texture as a surrogate marker of the risk to develop Estrogen Receptor (ER) sub-type specific breast cancer. A case-control study was performed, including 118 cancer cases stratified by ER receptor status and 354 age-matched controls. Digital mammographic (DM) images were retrospectively collected and analyzed under HIPAA and IRB approval. The performance of the texture features was compared to that of the standard mammographic density measures. We observed that breast percent density PD% and parenchymal texture features can both distinguish between cancer cases and controls (A z > 0.70). However, for ER subtype-specific classification, PD% alone does not provide sufficient classification (A z = 0.60), while texture features have significant classification performance (A z = 0.70). Combining breast density with texture features achieves the best performance (A z = 0.71). These findings suggest that mammographic texture analysis may have value for sub-type specific breast cancer risk assessment.",

author = "Gopal Karemore and Keller, {Brad M.} and Huen Oh and Julia Tchou and Mads Nielsen and Conant, {Emily F.} and Despina Kontos",

year = "2012",

doi = "10.1007/978-3-642-31271-7_77",

language = "English",

isbn = "978-3-642-31270-0",

series = "Lecture notes in computer science",

publisher = "Springer",

pages = "596--603",

editor = "Maidment, {Andrew D. A. } and Bakic, {Predrag R. } and Sara Gavenonis",

booktitle = "Breast Imaging",

note = "11th International Workshop on Breast Imaging ; Conference date: 08-07-2012 Through 11-07-2012",

}

TY - GEN

T1 - Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation

AU - Karemore, Gopal

AU - Keller, Brad M.

AU - Oh, Huen

AU - Tchou, Julia

AU - Nielsen, Mads

AU - Conant, Emily F.

AU - Kontos, Despina

N1 - Conference code: 11

PY - 2012

Y1 - 2012

N2 - We investigate the potential of mammographic parenchymal texture as a surrogate marker of the risk to develop Estrogen Receptor (ER) sub-type specific breast cancer. A case-control study was performed, including 118 cancer cases stratified by ER receptor status and 354 age-matched controls. Digital mammographic (DM) images were retrospectively collected and analyzed under HIPAA and IRB approval. The performance of the texture features was compared to that of the standard mammographic density measures. We observed that breast percent density PD% and parenchymal texture features can both distinguish between cancer cases and controls (A z > 0.70). However, for ER subtype-specific classification, PD% alone does not provide sufficient classification (A z = 0.60), while texture features have significant classification performance (A z = 0.70). Combining breast density with texture features achieves the best performance (A z = 0.71). These findings suggest that mammographic texture analysis may have value for sub-type specific breast cancer risk assessment.

AB - We investigate the potential of mammographic parenchymal texture as a surrogate marker of the risk to develop Estrogen Receptor (ER) sub-type specific breast cancer. A case-control study was performed, including 118 cancer cases stratified by ER receptor status and 354 age-matched controls. Digital mammographic (DM) images were retrospectively collected and analyzed under HIPAA and IRB approval. The performance of the texture features was compared to that of the standard mammographic density measures. We observed that breast percent density PD% and parenchymal texture features can both distinguish between cancer cases and controls (A z > 0.70). However, for ER subtype-specific classification, PD% alone does not provide sufficient classification (A z = 0.60), while texture features have significant classification performance (A z = 0.70). Combining breast density with texture features achieves the best performance (A z = 0.71). These findings suggest that mammographic texture analysis may have value for sub-type specific breast cancer risk assessment.

U2 - 10.1007/978-3-642-31271-7_77

DO - 10.1007/978-3-642-31271-7_77

M3 - Article in proceedings

SN - 978-3-642-31270-0

T3 - Lecture notes in computer science

SP - 596

EP - 603

BT - Breast Imaging

A2 - Maidment, Andrew D. A.

A2 - Bakic, Predrag R.

A2 - Gavenonis, Sara

PB - Springer

T2 - 11th International Workshop on Breast Imaging

Y2 - 8 July 2012 through 11 July 2012

ER -

Mammographic parenchymal texture analysis for estrogen-receptor subtype specific breast cancer risk estimation

Abstract

Workshop

UN SDGs

Access to Document

Fingerprint

Cite this