Computer-aided diagnosis of contrast-enhanced spectral mammography

A feasibility study

Bhavika K. Patel, Sara Ranjbar, Teresa Wu, Barbara A. Pockaj, Jing Li, Nan Zhang, Mark Lobbes, Bin Zhang, J. Ross Mitchell

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Objective To evaluate whether the use of a computer-aided diagnosis–contrast-enhanced spectral mammography (CAD-CESM) tool can further increase the diagnostic performance of CESM compared with that of experienced radiologists. Materials and methods This IRB-approved retrospective study analyzed 50 lesions described on CESM from August 2014 to December 2015. Histopathologic analyses, used as the criterion standard, revealed 24 benign and 26 malignant lesions. An expert breast radiologist manually outlined lesion boundaries on the different views. A set of morphologic and textural features were then extracted from the low-energy and recombined images. Machine-learning algorithms with feature selection were used along with statistical analysis to reduce, select, and combine features. Selected features were then used to construct a predictive model using a support vector machine (SVM) classification method in a leave-one-out–cross-validation approach. The classification performance was compared against the diagnostic predictions of 2 breast radiologists with access to the same CESM cases. Results Based on the SVM classification, CAD-CESM correctly identified 45 of 50 lesions in the cohort, resulting in an overall accuracy of 90%. The detection rate for the malignant group was 88% (3 false-negative cases) and 92% for the benign group (2 false-positive cases). Compared with the model, radiologist 1 had an overall accuracy of 78% and a detection rate of 92% (2 false-negative cases) for the malignant group and 62% (10 false-positive cases) for the benign group. Radiologist 2 had an overall accuracy of 86% and a detection rate of 100% for the malignant group and 71% (8 false-positive cases) for the benign group. Conclusions The results of our feasibility study suggest that a CAD-CESM tool can provide complementary information to radiologists, mainly by reducing the number of false-positive findings.

Original languageEnglish (US)
Pages (from-to)207-213
Number of pages7
JournalEuropean Journal of Radiology
Volume98
DOIs
StatePublished - Jan 1 2018

Fingerprint

Feasibility Studies
Mammography
Breast
Research Ethics Committees
Radiologists
Retrospective Studies

Keywords

  • Breast cancer
  • Computer-aided diagnosis
  • Contrast-enhanced digital mammography
  • Contrast-enhanced spectral mammography
  • Quantitative image analysis

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Computer-aided diagnosis of contrast-enhanced spectral mammography : A feasibility study. / Patel, Bhavika K.; Ranjbar, Sara; Wu, Teresa; Pockaj, Barbara A.; Li, Jing; Zhang, Nan; Lobbes, Mark; Zhang, Bin; Mitchell, J. Ross.

In: European Journal of Radiology, Vol. 98, 01.01.2018, p. 207-213.

Research output: Contribution to journalArticle

Patel, Bhavika K. ; Ranjbar, Sara ; Wu, Teresa ; Pockaj, Barbara A. ; Li, Jing ; Zhang, Nan ; Lobbes, Mark ; Zhang, Bin ; Mitchell, J. Ross. / Computer-aided diagnosis of contrast-enhanced spectral mammography : A feasibility study. In: European Journal of Radiology. 2018 ; Vol. 98. pp. 207-213.
@article{4cc451ef8d234760af7adbe3b1a322dd,
title = "Computer-aided diagnosis of contrast-enhanced spectral mammography: A feasibility study",
abstract = "Objective To evaluate whether the use of a computer-aided diagnosis–contrast-enhanced spectral mammography (CAD-CESM) tool can further increase the diagnostic performance of CESM compared with that of experienced radiologists. Materials and methods This IRB-approved retrospective study analyzed 50 lesions described on CESM from August 2014 to December 2015. Histopathologic analyses, used as the criterion standard, revealed 24 benign and 26 malignant lesions. An expert breast radiologist manually outlined lesion boundaries on the different views. A set of morphologic and textural features were then extracted from the low-energy and recombined images. Machine-learning algorithms with feature selection were used along with statistical analysis to reduce, select, and combine features. Selected features were then used to construct a predictive model using a support vector machine (SVM) classification method in a leave-one-out–cross-validation approach. The classification performance was compared against the diagnostic predictions of 2 breast radiologists with access to the same CESM cases. Results Based on the SVM classification, CAD-CESM correctly identified 45 of 50 lesions in the cohort, resulting in an overall accuracy of 90{\%}. The detection rate for the malignant group was 88{\%} (3 false-negative cases) and 92{\%} for the benign group (2 false-positive cases). Compared with the model, radiologist 1 had an overall accuracy of 78{\%} and a detection rate of 92{\%} (2 false-negative cases) for the malignant group and 62{\%} (10 false-positive cases) for the benign group. Radiologist 2 had an overall accuracy of 86{\%} and a detection rate of 100{\%} for the malignant group and 71{\%} (8 false-positive cases) for the benign group. Conclusions The results of our feasibility study suggest that a CAD-CESM tool can provide complementary information to radiologists, mainly by reducing the number of false-positive findings.",
keywords = "Breast cancer, Computer-aided diagnosis, Contrast-enhanced digital mammography, Contrast-enhanced spectral mammography, Quantitative image analysis",
author = "Patel, {Bhavika K.} and Sara Ranjbar and Teresa Wu and Pockaj, {Barbara A.} and Jing Li and Nan Zhang and Mark Lobbes and Bin Zhang and Mitchell, {J. Ross}",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.ejrad.2017.11.024",
language = "English (US)",
volume = "98",
pages = "207--213",
journal = "European Journal of Radiology",
issn = "0720-048X",
publisher = "Elsevier Ireland Ltd",

}

TY - JOUR

T1 - Computer-aided diagnosis of contrast-enhanced spectral mammography

T2 - A feasibility study

AU - Patel, Bhavika K.

AU - Ranjbar, Sara

AU - Wu, Teresa

AU - Pockaj, Barbara A.

AU - Li, Jing

AU - Zhang, Nan

AU - Lobbes, Mark

AU - Zhang, Bin

AU - Mitchell, J. Ross

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Objective To evaluate whether the use of a computer-aided diagnosis–contrast-enhanced spectral mammography (CAD-CESM) tool can further increase the diagnostic performance of CESM compared with that of experienced radiologists. Materials and methods This IRB-approved retrospective study analyzed 50 lesions described on CESM from August 2014 to December 2015. Histopathologic analyses, used as the criterion standard, revealed 24 benign and 26 malignant lesions. An expert breast radiologist manually outlined lesion boundaries on the different views. A set of morphologic and textural features were then extracted from the low-energy and recombined images. Machine-learning algorithms with feature selection were used along with statistical analysis to reduce, select, and combine features. Selected features were then used to construct a predictive model using a support vector machine (SVM) classification method in a leave-one-out–cross-validation approach. The classification performance was compared against the diagnostic predictions of 2 breast radiologists with access to the same CESM cases. Results Based on the SVM classification, CAD-CESM correctly identified 45 of 50 lesions in the cohort, resulting in an overall accuracy of 90%. The detection rate for the malignant group was 88% (3 false-negative cases) and 92% for the benign group (2 false-positive cases). Compared with the model, radiologist 1 had an overall accuracy of 78% and a detection rate of 92% (2 false-negative cases) for the malignant group and 62% (10 false-positive cases) for the benign group. Radiologist 2 had an overall accuracy of 86% and a detection rate of 100% for the malignant group and 71% (8 false-positive cases) for the benign group. Conclusions The results of our feasibility study suggest that a CAD-CESM tool can provide complementary information to radiologists, mainly by reducing the number of false-positive findings.

AB - Objective To evaluate whether the use of a computer-aided diagnosis–contrast-enhanced spectral mammography (CAD-CESM) tool can further increase the diagnostic performance of CESM compared with that of experienced radiologists. Materials and methods This IRB-approved retrospective study analyzed 50 lesions described on CESM from August 2014 to December 2015. Histopathologic analyses, used as the criterion standard, revealed 24 benign and 26 malignant lesions. An expert breast radiologist manually outlined lesion boundaries on the different views. A set of morphologic and textural features were then extracted from the low-energy and recombined images. Machine-learning algorithms with feature selection were used along with statistical analysis to reduce, select, and combine features. Selected features were then used to construct a predictive model using a support vector machine (SVM) classification method in a leave-one-out–cross-validation approach. The classification performance was compared against the diagnostic predictions of 2 breast radiologists with access to the same CESM cases. Results Based on the SVM classification, CAD-CESM correctly identified 45 of 50 lesions in the cohort, resulting in an overall accuracy of 90%. The detection rate for the malignant group was 88% (3 false-negative cases) and 92% for the benign group (2 false-positive cases). Compared with the model, radiologist 1 had an overall accuracy of 78% and a detection rate of 92% (2 false-negative cases) for the malignant group and 62% (10 false-positive cases) for the benign group. Radiologist 2 had an overall accuracy of 86% and a detection rate of 100% for the malignant group and 71% (8 false-positive cases) for the benign group. Conclusions The results of our feasibility study suggest that a CAD-CESM tool can provide complementary information to radiologists, mainly by reducing the number of false-positive findings.

KW - Breast cancer

KW - Computer-aided diagnosis

KW - Contrast-enhanced digital mammography

KW - Contrast-enhanced spectral mammography

KW - Quantitative image analysis

UR - http://www.scopus.com/inward/record.url?scp=85038874041&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038874041&partnerID=8YFLogxK

U2 - 10.1016/j.ejrad.2017.11.024

DO - 10.1016/j.ejrad.2017.11.024

M3 - Article

VL - 98

SP - 207

EP - 213

JO - European Journal of Radiology

JF - European Journal of Radiology

SN - 0720-048X

ER -