Additive Benefit of Radiomics Over Size Alone in the Distinction Between Benign Lesions and Luminal A Cancers on a Large Clinical Breast MRI Dataset

Heather M. Whitney*, Nathan S. Taylor, Karen Drukker, Alexandra V. Edwards, John Papaioannou, David V Schacht, Maryellen L. Giger

*Corresponding author for this work

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Rationale and Objectives: The objective of this study was to demonstrate improvement in distinguishing between benign lesions and luminal A breast cancers in a large clinical breast magnetic resonance imaging database by using quantitative radiomics over maximum linear size alone. Materials and Methods: In this retrospective study, 264 benign lesions and 390 luminal A breast cancers were automatically segmented from dynamic contrast-enhanced breast magnetic resonance images. Thirty-eight radiomic features were extracted. Tenfold cross validation was performed to assess the ability to distinguish between lesions and cancers using maximum linear size alone and lesion signatures obtained with stepwise feature selection and a linear discriminant analysis classifier including and excluding size features. Area under the receiver operating characteristic curve (AUC) was used as the figure of merit. Results: For maximum linear size alone, AUC and 95% confidence interval was 0.684 (0.642, 0.724) compared to 0.728 (0.687, 0.766) (P = 0.005) and 0.729 (0.689, 0.767) (P = 0.005) for lesion signature feature selection protocols including and excluding size features, respectively. The features of irregularity and entropy were chosen in all folds when size features were included and excluded. AUC for the radiomic signature using feature selection from all features was statistically equivalent to using feature selection from all features excluding size features, within an equivalence margin of 2%. Conclusions: Inclusion of multiple radiomic features, automatically extracted from magnetic resonance images, in a lesion signature significantly improved the ability to distinguish between benign lesions and luminal A breast cancers, compared to using maximum linear size alone. The radiomic features of irregularity and entropy appear to play an important but not a solitary role within the context of feature selection and computer-aided diagnosis.

Original languageEnglish (US)
Pages (from-to)202-209
Number of pages8
JournalAcademic Radiology
Volume26
Issue number2
DOIs
StatePublished - Feb 1 2019

Fingerprint

Area Under Curve
Breast
Entropy
Breast Neoplasms
Magnetic Resonance Spectroscopy
Neoplasms
Discriminant Analysis
ROC Curve
Retrospective Studies
Magnetic Resonance Imaging
Databases
Confidence Intervals
Datasets

Keywords

  • Breast cancer
  • MRI
  • luminal A
  • radiomics

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Whitney, Heather M. ; Taylor, Nathan S. ; Drukker, Karen ; Edwards, Alexandra V. ; Papaioannou, John ; Schacht, David V ; Giger, Maryellen L. / Additive Benefit of Radiomics Over Size Alone in the Distinction Between Benign Lesions and Luminal A Cancers on a Large Clinical Breast MRI Dataset. In: Academic Radiology. 2019 ; Vol. 26, No. 2. pp. 202-209.
@article{0613c698eb314a478de22713b15784b9,
title = "Additive Benefit of Radiomics Over Size Alone in the Distinction Between Benign Lesions and Luminal A Cancers on a Large Clinical Breast MRI Dataset",
abstract = "Rationale and Objectives: The objective of this study was to demonstrate improvement in distinguishing between benign lesions and luminal A breast cancers in a large clinical breast magnetic resonance imaging database by using quantitative radiomics over maximum linear size alone. Materials and Methods: In this retrospective study, 264 benign lesions and 390 luminal A breast cancers were automatically segmented from dynamic contrast-enhanced breast magnetic resonance images. Thirty-eight radiomic features were extracted. Tenfold cross validation was performed to assess the ability to distinguish between lesions and cancers using maximum linear size alone and lesion signatures obtained with stepwise feature selection and a linear discriminant analysis classifier including and excluding size features. Area under the receiver operating characteristic curve (AUC) was used as the figure of merit. Results: For maximum linear size alone, AUC and 95{\%} confidence interval was 0.684 (0.642, 0.724) compared to 0.728 (0.687, 0.766) (P = 0.005) and 0.729 (0.689, 0.767) (P = 0.005) for lesion signature feature selection protocols including and excluding size features, respectively. The features of irregularity and entropy were chosen in all folds when size features were included and excluded. AUC for the radiomic signature using feature selection from all features was statistically equivalent to using feature selection from all features excluding size features, within an equivalence margin of 2{\%}. Conclusions: Inclusion of multiple radiomic features, automatically extracted from magnetic resonance images, in a lesion signature significantly improved the ability to distinguish between benign lesions and luminal A breast cancers, compared to using maximum linear size alone. The radiomic features of irregularity and entropy appear to play an important but not a solitary role within the context of feature selection and computer-aided diagnosis.",
keywords = "Breast cancer, MRI, luminal A, radiomics",
author = "Whitney, {Heather M.} and Taylor, {Nathan S.} and Karen Drukker and Edwards, {Alexandra V.} and John Papaioannou and Schacht, {David V} and Giger, {Maryellen L.}",
year = "2019",
month = "2",
day = "1",
doi = "10.1016/j.acra.2018.04.019",
language = "English (US)",
volume = "26",
pages = "202--209",
journal = "Academic Radiology",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "2",

}

Additive Benefit of Radiomics Over Size Alone in the Distinction Between Benign Lesions and Luminal A Cancers on a Large Clinical Breast MRI Dataset. / Whitney, Heather M.; Taylor, Nathan S.; Drukker, Karen; Edwards, Alexandra V.; Papaioannou, John; Schacht, David V; Giger, Maryellen L.

In: Academic Radiology, Vol. 26, No. 2, 01.02.2019, p. 202-209.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Additive Benefit of Radiomics Over Size Alone in the Distinction Between Benign Lesions and Luminal A Cancers on a Large Clinical Breast MRI Dataset

AU - Whitney, Heather M.

AU - Taylor, Nathan S.

AU - Drukker, Karen

AU - Edwards, Alexandra V.

AU - Papaioannou, John

AU - Schacht, David V

AU - Giger, Maryellen L.

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Rationale and Objectives: The objective of this study was to demonstrate improvement in distinguishing between benign lesions and luminal A breast cancers in a large clinical breast magnetic resonance imaging database by using quantitative radiomics over maximum linear size alone. Materials and Methods: In this retrospective study, 264 benign lesions and 390 luminal A breast cancers were automatically segmented from dynamic contrast-enhanced breast magnetic resonance images. Thirty-eight radiomic features were extracted. Tenfold cross validation was performed to assess the ability to distinguish between lesions and cancers using maximum linear size alone and lesion signatures obtained with stepwise feature selection and a linear discriminant analysis classifier including and excluding size features. Area under the receiver operating characteristic curve (AUC) was used as the figure of merit. Results: For maximum linear size alone, AUC and 95% confidence interval was 0.684 (0.642, 0.724) compared to 0.728 (0.687, 0.766) (P = 0.005) and 0.729 (0.689, 0.767) (P = 0.005) for lesion signature feature selection protocols including and excluding size features, respectively. The features of irregularity and entropy were chosen in all folds when size features were included and excluded. AUC for the radiomic signature using feature selection from all features was statistically equivalent to using feature selection from all features excluding size features, within an equivalence margin of 2%. Conclusions: Inclusion of multiple radiomic features, automatically extracted from magnetic resonance images, in a lesion signature significantly improved the ability to distinguish between benign lesions and luminal A breast cancers, compared to using maximum linear size alone. The radiomic features of irregularity and entropy appear to play an important but not a solitary role within the context of feature selection and computer-aided diagnosis.

AB - Rationale and Objectives: The objective of this study was to demonstrate improvement in distinguishing between benign lesions and luminal A breast cancers in a large clinical breast magnetic resonance imaging database by using quantitative radiomics over maximum linear size alone. Materials and Methods: In this retrospective study, 264 benign lesions and 390 luminal A breast cancers were automatically segmented from dynamic contrast-enhanced breast magnetic resonance images. Thirty-eight radiomic features were extracted. Tenfold cross validation was performed to assess the ability to distinguish between lesions and cancers using maximum linear size alone and lesion signatures obtained with stepwise feature selection and a linear discriminant analysis classifier including and excluding size features. Area under the receiver operating characteristic curve (AUC) was used as the figure of merit. Results: For maximum linear size alone, AUC and 95% confidence interval was 0.684 (0.642, 0.724) compared to 0.728 (0.687, 0.766) (P = 0.005) and 0.729 (0.689, 0.767) (P = 0.005) for lesion signature feature selection protocols including and excluding size features, respectively. The features of irregularity and entropy were chosen in all folds when size features were included and excluded. AUC for the radiomic signature using feature selection from all features was statistically equivalent to using feature selection from all features excluding size features, within an equivalence margin of 2%. Conclusions: Inclusion of multiple radiomic features, automatically extracted from magnetic resonance images, in a lesion signature significantly improved the ability to distinguish between benign lesions and luminal A breast cancers, compared to using maximum linear size alone. The radiomic features of irregularity and entropy appear to play an important but not a solitary role within the context of feature selection and computer-aided diagnosis.

KW - Breast cancer

KW - MRI

KW - luminal A

KW - radiomics

UR - http://www.scopus.com/inward/record.url?scp=85046851541&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046851541&partnerID=8YFLogxK

U2 - 10.1016/j.acra.2018.04.019

DO - 10.1016/j.acra.2018.04.019

M3 - Article

C2 - 29754995

AN - SCOPUS:85046851541

VL - 26

SP - 202

EP - 209

JO - Academic Radiology

JF - Academic Radiology

SN - 1076-6332

IS - 2

ER -