Breast cancer Ki-67 expression prediction by digital breast tomosynthesis radiomics features

Alberto Stefano Tagliafico, Bianca Bignotti, Federica Rossi, Joao Matos, Massimo Calabrese, Francesca Valdora, Nehmat Houssami

Research output: Contribution to journalArticle

Abstract

BACKGROUND: To investigate whether quantitative radiomic features extracted from digital breast tomosynthesis (DBT) are associated with Ki-67 expression of breast cancer. MATERIALS AND METHODS: This is a prospective ethically approved study of 70 women diagnosed with invasive breast cancer in 2018, including 40 low Ki-67 expression (Ki-67 proliferation index <14%) cases and 30 high Ki-67 expression (Ki-67 proliferation index ≥ 14%) cases. A set of 106 quantitative radiomic features, including morphological, grey/scale statistics, and texture features, were extracted from DBT images. After applying least absolute shrinkage and selection operator (LASSO) method to select the most predictive features set for the classifiers, low versus high Ki-67 expression was evaluated by the area under the curve (AUC) at receiver operating characteristic analysis. Correlation coefficient was calculated for the most significant features. RESULTS: A combination of five features yielded AUC of up to 0.698. The five most predictive features (sphericity, autocorrelation, interquartile range, robust mean absolute deviation, and short-run high grey-level emphasis) showed a statistical significance (p ≤ 0.001) in the classification. Thirty-four features were significantly (p ≤ 0.001) correlated with Ki-67, and five of these had a correlation coefficient of > 0.5. CONCLUSION: The present study showed that quantitative radiomic imaging features of breast tumour extracted from DBT images are associated with breast cancer Ki-67 expression. Larger studies are needed in order to further evaluate these findings.

Original languageEnglish
Number of pages1
JournalEuropean radiology experimental
Volume3
Issue number1
DOIs
Publication statusPublished - Aug 14 2019

Fingerprint

Mammography
Breast Neoplasms

Keywords

  • Breast neoplasms
  • Cell proliferation
  • Ki-67 expression
  • Mammography
  • Radiomics

Cite this

Breast cancer Ki-67 expression prediction by digital breast tomosynthesis radiomics features. / Tagliafico, Alberto Stefano; Bignotti, Bianca; Rossi, Federica; Matos, Joao; Calabrese, Massimo; Valdora, Francesca; Houssami, Nehmat.

In: European radiology experimental, Vol. 3, No. 1, 14.08.2019.

Research output: Contribution to journalArticle

Tagliafico, Alberto Stefano ; Bignotti, Bianca ; Rossi, Federica ; Matos, Joao ; Calabrese, Massimo ; Valdora, Francesca ; Houssami, Nehmat. / Breast cancer Ki-67 expression prediction by digital breast tomosynthesis radiomics features. In: European radiology experimental. 2019 ; Vol. 3, No. 1.
@article{8492ea7363144c8eb986f380eae2cabb,
title = "Breast cancer Ki-67 expression prediction by digital breast tomosynthesis radiomics features",
abstract = "BACKGROUND: To investigate whether quantitative radiomic features extracted from digital breast tomosynthesis (DBT) are associated with Ki-67 expression of breast cancer. MATERIALS AND METHODS: This is a prospective ethically approved study of 70 women diagnosed with invasive breast cancer in 2018, including 40 low Ki-67 expression (Ki-67 proliferation index <14{\%}) cases and 30 high Ki-67 expression (Ki-67 proliferation index ≥ 14{\%}) cases. A set of 106 quantitative radiomic features, including morphological, grey/scale statistics, and texture features, were extracted from DBT images. After applying least absolute shrinkage and selection operator (LASSO) method to select the most predictive features set for the classifiers, low versus high Ki-67 expression was evaluated by the area under the curve (AUC) at receiver operating characteristic analysis. Correlation coefficient was calculated for the most significant features. RESULTS: A combination of five features yielded AUC of up to 0.698. The five most predictive features (sphericity, autocorrelation, interquartile range, robust mean absolute deviation, and short-run high grey-level emphasis) showed a statistical significance (p ≤ 0.001) in the classification. Thirty-four features were significantly (p ≤ 0.001) correlated with Ki-67, and five of these had a correlation coefficient of > 0.5. CONCLUSION: The present study showed that quantitative radiomic imaging features of breast tumour extracted from DBT images are associated with breast cancer Ki-67 expression. Larger studies are needed in order to further evaluate these findings.",
keywords = "Breast neoplasms, Cell proliferation, Ki-67 expression, Mammography, Radiomics",
author = "Tagliafico, {Alberto Stefano} and Bianca Bignotti and Federica Rossi and Joao Matos and Massimo Calabrese and Francesca Valdora and Nehmat Houssami",
year = "2019",
month = "8",
day = "14",
doi = "10.1186/s41747-019-0117-2",
language = "English",
volume = "3",
journal = "European radiology experimental",
issn = "2509-9280",
publisher = "Springer Open",
number = "1",

}

TY - JOUR

T1 - Breast cancer Ki-67 expression prediction by digital breast tomosynthesis radiomics features

AU - Tagliafico, Alberto Stefano

AU - Bignotti, Bianca

AU - Rossi, Federica

AU - Matos, Joao

AU - Calabrese, Massimo

AU - Valdora, Francesca

AU - Houssami, Nehmat

PY - 2019/8/14

Y1 - 2019/8/14

N2 - BACKGROUND: To investigate whether quantitative radiomic features extracted from digital breast tomosynthesis (DBT) are associated with Ki-67 expression of breast cancer. MATERIALS AND METHODS: This is a prospective ethically approved study of 70 women diagnosed with invasive breast cancer in 2018, including 40 low Ki-67 expression (Ki-67 proliferation index <14%) cases and 30 high Ki-67 expression (Ki-67 proliferation index ≥ 14%) cases. A set of 106 quantitative radiomic features, including morphological, grey/scale statistics, and texture features, were extracted from DBT images. After applying least absolute shrinkage and selection operator (LASSO) method to select the most predictive features set for the classifiers, low versus high Ki-67 expression was evaluated by the area under the curve (AUC) at receiver operating characteristic analysis. Correlation coefficient was calculated for the most significant features. RESULTS: A combination of five features yielded AUC of up to 0.698. The five most predictive features (sphericity, autocorrelation, interquartile range, robust mean absolute deviation, and short-run high grey-level emphasis) showed a statistical significance (p ≤ 0.001) in the classification. Thirty-four features were significantly (p ≤ 0.001) correlated with Ki-67, and five of these had a correlation coefficient of > 0.5. CONCLUSION: The present study showed that quantitative radiomic imaging features of breast tumour extracted from DBT images are associated with breast cancer Ki-67 expression. Larger studies are needed in order to further evaluate these findings.

AB - BACKGROUND: To investigate whether quantitative radiomic features extracted from digital breast tomosynthesis (DBT) are associated with Ki-67 expression of breast cancer. MATERIALS AND METHODS: This is a prospective ethically approved study of 70 women diagnosed with invasive breast cancer in 2018, including 40 low Ki-67 expression (Ki-67 proliferation index <14%) cases and 30 high Ki-67 expression (Ki-67 proliferation index ≥ 14%) cases. A set of 106 quantitative radiomic features, including morphological, grey/scale statistics, and texture features, were extracted from DBT images. After applying least absolute shrinkage and selection operator (LASSO) method to select the most predictive features set for the classifiers, low versus high Ki-67 expression was evaluated by the area under the curve (AUC) at receiver operating characteristic analysis. Correlation coefficient was calculated for the most significant features. RESULTS: A combination of five features yielded AUC of up to 0.698. The five most predictive features (sphericity, autocorrelation, interquartile range, robust mean absolute deviation, and short-run high grey-level emphasis) showed a statistical significance (p ≤ 0.001) in the classification. Thirty-four features were significantly (p ≤ 0.001) correlated with Ki-67, and five of these had a correlation coefficient of > 0.5. CONCLUSION: The present study showed that quantitative radiomic imaging features of breast tumour extracted from DBT images are associated with breast cancer Ki-67 expression. Larger studies are needed in order to further evaluate these findings.

KW - Breast neoplasms

KW - Cell proliferation

KW - Ki-67 expression

KW - Mammography

KW - Radiomics

UR - http://www.scopus.com/inward/record.url?scp=85071281306&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071281306&partnerID=8YFLogxK

U2 - 10.1186/s41747-019-0117-2

DO - 10.1186/s41747-019-0117-2

M3 - Article

VL - 3

JO - European radiology experimental

JF - European radiology experimental

SN - 2509-9280

IS - 1

ER -