Computerized detection of breast tissue asymmetry depicted on bilateral mammograms. A preliminary study of breast risk stratification

Xingwei Wang, Dror Lederman, Jun Tan, Xiao Hui Wang, Bin Zheng

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

Rationale and Objectives: Assessment of the breast tissue pattern asymmetry depicted on bilateral mammograms is routinely used by radiologists when reading and interpreting mammograms. The purpose of this study is to develop an automated scheme to detect breast tissue asymmetry depicted on bilateral mammograms and use the computed asymmetric features to predict the likelihood (or the risk) of women having or developing breast abnormalities or cancer. Materials and Methods: A testing dataset was selected from a large and diverse full-field digital mammography image database, which includes 100 randomly selected negative cases (not recalled during the screening) and 100 positive cases for having or developing breast abnormalities or cancer. Among these positive cases 40 were recalled (biopsy) because of suspicious findings in which 8 were determined as high risk with the lesions surgically removed and the remaining were proven to be benign, and 60 cases were acquired from examinations that were interpreted as negative (without dominant masses or microcalcifications) but the cancers were detected 6-18 months later. A computerized scheme was developed to detect asymmetry of mammographic tissue density represented by the related feature differences computed from bilateral images. Initially, each of 20 features was tested to classify between the positive and the negative cases. To further improve the classification performance, a genetic algorithm (GA) was applied to select a set of optimal features and build an artificial neural network (ANN). The leave-one-case-out validation method was used to evaluate the ANN classification performance. Results: Using a single feature, the maximum classification performance level measured by the area under the receiver operating characteristic curve (AUC) was 0.681 ± 0.038. Using the GA-optimized ANN, the classification performance level increased to an AUC = 0.754 ± 0.024. At 90% specificity, the ANN classifier yielded 42% sensitivity, in which 42 positive cases were correctly identified. Among them, 30 were the " prior" examinations of the cancer cases and 12 were recalled benign cases, which represent 50% and 30% sensitivity levels in these two subgroups, respectively. Conclusions: This study demonstrated that using the computerized detected feature differences related to the bilateral mammographic breast tissue asymmetry, an automated scheme is able to classify a set of testing cases into the two groups of positive or negative of having or developing breast abnormalities or cancer. Hence, further development and optimization of this automated method may eventually help radiologists identify a fraction of women at high risk of developing breast cancer and ultimately detect cancer at an early stage.

Original languageEnglish (US)
Pages (from-to)1234-1241
Number of pages8
JournalAcademic Radiology
Volume17
Issue number10
DOIs
StatePublished - Oct 1 2010

Fingerprint

Breast
Neoplasms
Area Under Curve
Calcinosis
Mammography
ROC Curve
Reading
Databases
Breast Neoplasms
Biopsy
Radiologists

Keywords

  • Breast cancer
  • Computer-aided detection (CAD)
  • Mammographic breast tissue asymmetry
  • Mammography
  • Risk assessment

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Computerized detection of breast tissue asymmetry depicted on bilateral mammograms. A preliminary study of breast risk stratification. / Wang, Xingwei; Lederman, Dror; Tan, Jun; Wang, Xiao Hui; Zheng, Bin.

In: Academic Radiology, Vol. 17, No. 10, 01.10.2010, p. 1234-1241.

Research output: Contribution to journalArticle

Wang, Xingwei ; Lederman, Dror ; Tan, Jun ; Wang, Xiao Hui ; Zheng, Bin. / Computerized detection of breast tissue asymmetry depicted on bilateral mammograms. A preliminary study of breast risk stratification. In: Academic Radiology. 2010 ; Vol. 17, No. 10. pp. 1234-1241.
@article{b21052d6f166476880ce1095e22798e0,
title = "Computerized detection of breast tissue asymmetry depicted on bilateral mammograms. A preliminary study of breast risk stratification",
abstract = "Rationale and Objectives: Assessment of the breast tissue pattern asymmetry depicted on bilateral mammograms is routinely used by radiologists when reading and interpreting mammograms. The purpose of this study is to develop an automated scheme to detect breast tissue asymmetry depicted on bilateral mammograms and use the computed asymmetric features to predict the likelihood (or the risk) of women having or developing breast abnormalities or cancer. Materials and Methods: A testing dataset was selected from a large and diverse full-field digital mammography image database, which includes 100 randomly selected negative cases (not recalled during the screening) and 100 positive cases for having or developing breast abnormalities or cancer. Among these positive cases 40 were recalled (biopsy) because of suspicious findings in which 8 were determined as high risk with the lesions surgically removed and the remaining were proven to be benign, and 60 cases were acquired from examinations that were interpreted as negative (without dominant masses or microcalcifications) but the cancers were detected 6-18 months later. A computerized scheme was developed to detect asymmetry of mammographic tissue density represented by the related feature differences computed from bilateral images. Initially, each of 20 features was tested to classify between the positive and the negative cases. To further improve the classification performance, a genetic algorithm (GA) was applied to select a set of optimal features and build an artificial neural network (ANN). The leave-one-case-out validation method was used to evaluate the ANN classification performance. Results: Using a single feature, the maximum classification performance level measured by the area under the receiver operating characteristic curve (AUC) was 0.681 ± 0.038. Using the GA-optimized ANN, the classification performance level increased to an AUC = 0.754 ± 0.024. At 90{\%} specificity, the ANN classifier yielded 42{\%} sensitivity, in which 42 positive cases were correctly identified. Among them, 30 were the {"} prior{"} examinations of the cancer cases and 12 were recalled benign cases, which represent 50{\%} and 30{\%} sensitivity levels in these two subgroups, respectively. Conclusions: This study demonstrated that using the computerized detected feature differences related to the bilateral mammographic breast tissue asymmetry, an automated scheme is able to classify a set of testing cases into the two groups of positive or negative of having or developing breast abnormalities or cancer. Hence, further development and optimization of this automated method may eventually help radiologists identify a fraction of women at high risk of developing breast cancer and ultimately detect cancer at an early stage.",
keywords = "Breast cancer, Computer-aided detection (CAD), Mammographic breast tissue asymmetry, Mammography, Risk assessment",
author = "Xingwei Wang and Dror Lederman and Jun Tan and Wang, {Xiao Hui} and Bin Zheng",
year = "2010",
month = "10",
day = "1",
doi = "10.1016/j.acra.2010.05.016",
language = "English (US)",
volume = "17",
pages = "1234--1241",
journal = "Academic Radiology",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "10",

}

TY - JOUR

T1 - Computerized detection of breast tissue asymmetry depicted on bilateral mammograms. A preliminary study of breast risk stratification

AU - Wang, Xingwei

AU - Lederman, Dror

AU - Tan, Jun

AU - Wang, Xiao Hui

AU - Zheng, Bin

PY - 2010/10/1

Y1 - 2010/10/1

N2 - Rationale and Objectives: Assessment of the breast tissue pattern asymmetry depicted on bilateral mammograms is routinely used by radiologists when reading and interpreting mammograms. The purpose of this study is to develop an automated scheme to detect breast tissue asymmetry depicted on bilateral mammograms and use the computed asymmetric features to predict the likelihood (or the risk) of women having or developing breast abnormalities or cancer. Materials and Methods: A testing dataset was selected from a large and diverse full-field digital mammography image database, which includes 100 randomly selected negative cases (not recalled during the screening) and 100 positive cases for having or developing breast abnormalities or cancer. Among these positive cases 40 were recalled (biopsy) because of suspicious findings in which 8 were determined as high risk with the lesions surgically removed and the remaining were proven to be benign, and 60 cases were acquired from examinations that were interpreted as negative (without dominant masses or microcalcifications) but the cancers were detected 6-18 months later. A computerized scheme was developed to detect asymmetry of mammographic tissue density represented by the related feature differences computed from bilateral images. Initially, each of 20 features was tested to classify between the positive and the negative cases. To further improve the classification performance, a genetic algorithm (GA) was applied to select a set of optimal features and build an artificial neural network (ANN). The leave-one-case-out validation method was used to evaluate the ANN classification performance. Results: Using a single feature, the maximum classification performance level measured by the area under the receiver operating characteristic curve (AUC) was 0.681 ± 0.038. Using the GA-optimized ANN, the classification performance level increased to an AUC = 0.754 ± 0.024. At 90% specificity, the ANN classifier yielded 42% sensitivity, in which 42 positive cases were correctly identified. Among them, 30 were the " prior" examinations of the cancer cases and 12 were recalled benign cases, which represent 50% and 30% sensitivity levels in these two subgroups, respectively. Conclusions: This study demonstrated that using the computerized detected feature differences related to the bilateral mammographic breast tissue asymmetry, an automated scheme is able to classify a set of testing cases into the two groups of positive or negative of having or developing breast abnormalities or cancer. Hence, further development and optimization of this automated method may eventually help radiologists identify a fraction of women at high risk of developing breast cancer and ultimately detect cancer at an early stage.

AB - Rationale and Objectives: Assessment of the breast tissue pattern asymmetry depicted on bilateral mammograms is routinely used by radiologists when reading and interpreting mammograms. The purpose of this study is to develop an automated scheme to detect breast tissue asymmetry depicted on bilateral mammograms and use the computed asymmetric features to predict the likelihood (or the risk) of women having or developing breast abnormalities or cancer. Materials and Methods: A testing dataset was selected from a large and diverse full-field digital mammography image database, which includes 100 randomly selected negative cases (not recalled during the screening) and 100 positive cases for having or developing breast abnormalities or cancer. Among these positive cases 40 were recalled (biopsy) because of suspicious findings in which 8 were determined as high risk with the lesions surgically removed and the remaining were proven to be benign, and 60 cases were acquired from examinations that were interpreted as negative (without dominant masses or microcalcifications) but the cancers were detected 6-18 months later. A computerized scheme was developed to detect asymmetry of mammographic tissue density represented by the related feature differences computed from bilateral images. Initially, each of 20 features was tested to classify between the positive and the negative cases. To further improve the classification performance, a genetic algorithm (GA) was applied to select a set of optimal features and build an artificial neural network (ANN). The leave-one-case-out validation method was used to evaluate the ANN classification performance. Results: Using a single feature, the maximum classification performance level measured by the area under the receiver operating characteristic curve (AUC) was 0.681 ± 0.038. Using the GA-optimized ANN, the classification performance level increased to an AUC = 0.754 ± 0.024. At 90% specificity, the ANN classifier yielded 42% sensitivity, in which 42 positive cases were correctly identified. Among them, 30 were the " prior" examinations of the cancer cases and 12 were recalled benign cases, which represent 50% and 30% sensitivity levels in these two subgroups, respectively. Conclusions: This study demonstrated that using the computerized detected feature differences related to the bilateral mammographic breast tissue asymmetry, an automated scheme is able to classify a set of testing cases into the two groups of positive or negative of having or developing breast abnormalities or cancer. Hence, further development and optimization of this automated method may eventually help radiologists identify a fraction of women at high risk of developing breast cancer and ultimately detect cancer at an early stage.

KW - Breast cancer

KW - Computer-aided detection (CAD)

KW - Mammographic breast tissue asymmetry

KW - Mammography

KW - Risk assessment

UR - http://www.scopus.com/inward/record.url?scp=77956482390&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956482390&partnerID=8YFLogxK

U2 - 10.1016/j.acra.2010.05.016

DO - 10.1016/j.acra.2010.05.016

M3 - Article

VL - 17

SP - 1234

EP - 1241

JO - Academic Radiology

JF - Academic Radiology

SN - 1076-6332

IS - 10

ER -