Classification of Breast Masses Using a Computer-Aided Diagnosis Scheme of Contrast Enhanced Digital Mammograms

Gopichandh Danala, Bhavika Patel, Faranak Aghaei, Morteza Heidari, Jing Li, Teresa Wu, Bin Zheng

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Contrast-enhanced digital mammography (CEDM) is a promising imaging modality in breast cancer diagnosis. This study aims to investigate how to optimally develop a computer-aided diagnosis (CAD) scheme of CEDM images to classify breast masses. A CEDM dataset of 111 patients was assembled, which includes 33 benign and 78 malignant cases. Each CEDM includes two types of images namely, low energy (LE) and dual-energy subtracted (DES) images. A CAD scheme was applied to segment mass regions depicting on LE and DES images separately. Optimal segmentation results generated from DES images were also mapped to LE images or vice versa. After computing image features, multilayer perceptron based machine learning classifiers that integrate with a correlation-based feature subset evaluator and leave-one-case-out cross-validation method were built to classify mass regions. When applying CAD to DES and LE images with original segmentation, areas under ROC curves (AUC) were 0.759 ± 0.053 and 0.753 ± 0.047, respectively. After mapping the mass regions optimally segmented on DES images to LE images, AUC significantly increased to 0.848 ± 0.038 (p < 0.01). Study demonstrated that DES images eliminated overlapping effect of dense breast tissue, which helps improve mass segmentation accuracy. The study demonstrated that applying a novel approach to optimally map mass region segmented from DES images to LE images enabled CAD to yield significantly improved performance.

Original languageEnglish (US)
Pages (from-to)1-13
Number of pages13
JournalAnnals of Biomedical Engineering
DOIs
StateAccepted/In press - May 10 2018

Fingerprint

Computer aided diagnosis
Mammography
Multilayer neural networks
Learning systems
Classifiers
Tissue
Imaging techniques

Keywords

  • Breast cancer diagnosis
  • Classification of breast masses
  • Computer-aided diagnosis (CAD)
  • Contrast-enhanced digital mammography (CEDM)
  • Performance comparison
  • Segmentation of breast mass regions

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Classification of Breast Masses Using a Computer-Aided Diagnosis Scheme of Contrast Enhanced Digital Mammograms. / Danala, Gopichandh; Patel, Bhavika; Aghaei, Faranak; Heidari, Morteza; Li, Jing; Wu, Teresa; Zheng, Bin.

In: Annals of Biomedical Engineering, 10.05.2018, p. 1-13.

Research output: Contribution to journalArticle

Danala, Gopichandh ; Patel, Bhavika ; Aghaei, Faranak ; Heidari, Morteza ; Li, Jing ; Wu, Teresa ; Zheng, Bin. / Classification of Breast Masses Using a Computer-Aided Diagnosis Scheme of Contrast Enhanced Digital Mammograms. In: Annals of Biomedical Engineering. 2018 ; pp. 1-13.
@article{a9037bd7d0714a78ac8d72f07a8fbbf4,
title = "Classification of Breast Masses Using a Computer-Aided Diagnosis Scheme of Contrast Enhanced Digital Mammograms",
abstract = "Contrast-enhanced digital mammography (CEDM) is a promising imaging modality in breast cancer diagnosis. This study aims to investigate how to optimally develop a computer-aided diagnosis (CAD) scheme of CEDM images to classify breast masses. A CEDM dataset of 111 patients was assembled, which includes 33 benign and 78 malignant cases. Each CEDM includes two types of images namely, low energy (LE) and dual-energy subtracted (DES) images. A CAD scheme was applied to segment mass regions depicting on LE and DES images separately. Optimal segmentation results generated from DES images were also mapped to LE images or vice versa. After computing image features, multilayer perceptron based machine learning classifiers that integrate with a correlation-based feature subset evaluator and leave-one-case-out cross-validation method were built to classify mass regions. When applying CAD to DES and LE images with original segmentation, areas under ROC curves (AUC) were 0.759 ± 0.053 and 0.753 ± 0.047, respectively. After mapping the mass regions optimally segmented on DES images to LE images, AUC significantly increased to 0.848 ± 0.038 (p < 0.01). Study demonstrated that DES images eliminated overlapping effect of dense breast tissue, which helps improve mass segmentation accuracy. The study demonstrated that applying a novel approach to optimally map mass region segmented from DES images to LE images enabled CAD to yield significantly improved performance.",
keywords = "Breast cancer diagnosis, Classification of breast masses, Computer-aided diagnosis (CAD), Contrast-enhanced digital mammography (CEDM), Performance comparison, Segmentation of breast mass regions",
author = "Gopichandh Danala and Bhavika Patel and Faranak Aghaei and Morteza Heidari and Jing Li and Teresa Wu and Bin Zheng",
year = "2018",
month = "5",
day = "10",
doi = "10.1007/s10439-018-2044-4",
language = "English (US)",
pages = "1--13",
journal = "Annals of Biomedical Engineering",
issn = "0090-6964",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Classification of Breast Masses Using a Computer-Aided Diagnosis Scheme of Contrast Enhanced Digital Mammograms

AU - Danala, Gopichandh

AU - Patel, Bhavika

AU - Aghaei, Faranak

AU - Heidari, Morteza

AU - Li, Jing

AU - Wu, Teresa

AU - Zheng, Bin

PY - 2018/5/10

Y1 - 2018/5/10

N2 - Contrast-enhanced digital mammography (CEDM) is a promising imaging modality in breast cancer diagnosis. This study aims to investigate how to optimally develop a computer-aided diagnosis (CAD) scheme of CEDM images to classify breast masses. A CEDM dataset of 111 patients was assembled, which includes 33 benign and 78 malignant cases. Each CEDM includes two types of images namely, low energy (LE) and dual-energy subtracted (DES) images. A CAD scheme was applied to segment mass regions depicting on LE and DES images separately. Optimal segmentation results generated from DES images were also mapped to LE images or vice versa. After computing image features, multilayer perceptron based machine learning classifiers that integrate with a correlation-based feature subset evaluator and leave-one-case-out cross-validation method were built to classify mass regions. When applying CAD to DES and LE images with original segmentation, areas under ROC curves (AUC) were 0.759 ± 0.053 and 0.753 ± 0.047, respectively. After mapping the mass regions optimally segmented on DES images to LE images, AUC significantly increased to 0.848 ± 0.038 (p < 0.01). Study demonstrated that DES images eliminated overlapping effect of dense breast tissue, which helps improve mass segmentation accuracy. The study demonstrated that applying a novel approach to optimally map mass region segmented from DES images to LE images enabled CAD to yield significantly improved performance.

AB - Contrast-enhanced digital mammography (CEDM) is a promising imaging modality in breast cancer diagnosis. This study aims to investigate how to optimally develop a computer-aided diagnosis (CAD) scheme of CEDM images to classify breast masses. A CEDM dataset of 111 patients was assembled, which includes 33 benign and 78 malignant cases. Each CEDM includes two types of images namely, low energy (LE) and dual-energy subtracted (DES) images. A CAD scheme was applied to segment mass regions depicting on LE and DES images separately. Optimal segmentation results generated from DES images were also mapped to LE images or vice versa. After computing image features, multilayer perceptron based machine learning classifiers that integrate with a correlation-based feature subset evaluator and leave-one-case-out cross-validation method were built to classify mass regions. When applying CAD to DES and LE images with original segmentation, areas under ROC curves (AUC) were 0.759 ± 0.053 and 0.753 ± 0.047, respectively. After mapping the mass regions optimally segmented on DES images to LE images, AUC significantly increased to 0.848 ± 0.038 (p < 0.01). Study demonstrated that DES images eliminated overlapping effect of dense breast tissue, which helps improve mass segmentation accuracy. The study demonstrated that applying a novel approach to optimally map mass region segmented from DES images to LE images enabled CAD to yield significantly improved performance.

KW - Breast cancer diagnosis

KW - Classification of breast masses

KW - Computer-aided diagnosis (CAD)

KW - Contrast-enhanced digital mammography (CEDM)

KW - Performance comparison

KW - Segmentation of breast mass regions

UR - http://www.scopus.com/inward/record.url?scp=85046754828&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046754828&partnerID=8YFLogxK

U2 - 10.1007/s10439-018-2044-4

DO - 10.1007/s10439-018-2044-4

M3 - Article

SP - 1

EP - 13

JO - Annals of Biomedical Engineering

JF - Annals of Biomedical Engineering

SN - 0090-6964

ER -