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

Authors:
Gopal Karemore;Brad M. Keller;Huen Oh;Julia Tchou;Mads Nielsen;Emily F. Conant;Despina Kontos
Affiliations:
University of Pennsylvania, PA, USA,University of Copenhagen, Copenhagen, Denmark;University of Pennsylvania, PA;University of Pennsylvania, PA;University of Pennsylvania, PA;University of Copenhagen, Copenhagen, Denmark;University of Pennsylvania, PA;University of Pennsylvania, PA
Venue:
IWDM'12 Proceedings of the 11th international conference on Breast Imaging
Year:
2012

Citing 5
Cited 3

An optimal algorithm for approximate nearest neighbor searching fixed dimensions

Journal of the ACM (JACM)
Coherence-Enhancing Diffusion Filtering

International Journal of Computer Vision
Feature selection for ranking

SIGIR '07 Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval
Assessing estrogen receptors' status by texture analysis of breast tissue specimens and pattern recognition methods

CAIP'07 Proceedings of the 12th international conference on Computer analysis of images and patterns
Adaptive multi-cluster fuzzy C-means segmentation of breast parenchymal tissue in digital mammography

MICCAI'11 Proceedings of the 14th international conference on Medical image computing and computer-assisted intervention - Volume Part III

Background intensity independent texture features for assessing breast cancer risk in screening mammograms

Pattern Recognition Letters
Constructing and applying higher order textons: Estimating breast cancer risk

Pattern Recognition
Texture and region dependent breast cancer risk assessment from screening mammograms

Pattern Recognition Letters

Quantified Score

Hi-index	0.00

Visualization

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 (Az 0.70). However, for ER subtype-specific classification, PD% alone does not provide sufficient classification (Az = 0.60), while texture features have significant classification performance (Az = 0.70). Combining breast density with texture features achieves the best performance (Az = 0.71). These findings suggest that mammographic texture analysis may have value for sub-type specific breast cancer risk assessment.