Magnetic resonance imaging texture analysis classification of primary breast cancer

S. A. Waugh (Lead / Corresponding author), C. A. Purdie, L. B. Jordan, S. Vinnicombe, R. A. Lerski, P. Martin, A. M. Thompson

    Research output: Contribution to journalArticle

    60 Citations (Scopus)

    Abstract

    Objectives: Patient-tailored treatments for breast cancer are based on histological and immunohistochemical (IHC) subtypes. Magnetic Resonance Imaging (MRI) texture analysis (TA) may be useful in non-invasive lesion subtype classification. Methods: Women with newly diagnosed primary breast cancer underwent pre-treatment dynamic contrast-enhanced breast MRI. TA was performed using co-occurrence matrix (COM) features, by creating a model on retrospective training data, then prospectively applying to a test set. Analyses were blinded to breast pathology. Subtype classifications were performed using a cross-validated k-nearest-neighbour (k = 3) technique, with accuracy relative to pathology assessed and receiver operator curve (AUROC) calculated. Mann-Whitney U and Kruskal-Wallis tests were used to assess raw entropy feature values. Results: Histological subtype classifications were similar across training (n = 148 cancers) and test sets (n = 73 lesions) using all COM features (training: 75 %, AUROC = 0.816; test: 72.5 %, AUROC = 0.823). Entropy features were significantly different between lobular and ductal cancers (p < 0.001; Mann-Whitney U). IHC classifications using COM features were also similar for training and test data (training: 57.2 %, AUROC = 0.754; test: 57.0 %, AUROC = 0.750). Hormone receptor positive and negative cancers demonstrated significantly different entropy features. Entropy features alone were unable to create a robust classification model. Conclusion: Textural differences on contrast-enhanced MR images may reflect underlying lesion subtypes, which merits testing against treatment response. 

    Original languageEnglish
    Pages (from-to)322-330
    Number of pages9
    JournalEuropean Radiology
    Volume26
    Issue number2
    Early online date12 Jun 2015
    DOIs
    Publication statusPublished - Feb 2016

    Fingerprint

    Entropy
    Magnetic Resonance Imaging
    Breast Neoplasms
    Breast
    Pathology
    Neoplasms
    Therapeutics
    Hormones

    Keywords

    • Breast cancer
    • Classification
    • Histological subtypes and immunohistochemical profiles
    • Magnetic Resonance Imaging (MRI)
    • Texture analysis (TA)

    Cite this

    Waugh, S. A., Purdie, C. A., Jordan, L. B., Vinnicombe, S., Lerski, R. A., Martin, P., & Thompson, A. M. (2016). Magnetic resonance imaging texture analysis classification of primary breast cancer. European Radiology, 26(2), 322-330. https://doi.org/10.1007/s00330-015-3845-6
    Waugh, S. A. ; Purdie, C. A. ; Jordan, L. B. ; Vinnicombe, S. ; Lerski, R. A. ; Martin, P. ; Thompson, A. M. / Magnetic resonance imaging texture analysis classification of primary breast cancer. In: European Radiology. 2016 ; Vol. 26, No. 2. pp. 322-330.
    @article{d6137925b36a438ea098131943ac56f6,
    title = "Magnetic resonance imaging texture analysis classification of primary breast cancer",
    abstract = "Objectives: Patient-tailored treatments for breast cancer are based on histological and immunohistochemical (IHC) subtypes. Magnetic Resonance Imaging (MRI) texture analysis (TA) may be useful in non-invasive lesion subtype classification. Methods: Women with newly diagnosed primary breast cancer underwent pre-treatment dynamic contrast-enhanced breast MRI. TA was performed using co-occurrence matrix (COM) features, by creating a model on retrospective training data, then prospectively applying to a test set. Analyses were blinded to breast pathology. Subtype classifications were performed using a cross-validated k-nearest-neighbour (k = 3) technique, with accuracy relative to pathology assessed and receiver operator curve (AUROC) calculated. Mann-Whitney U and Kruskal-Wallis tests were used to assess raw entropy feature values. Results: Histological subtype classifications were similar across training (n = 148 cancers) and test sets (n = 73 lesions) using all COM features (training: 75 {\%}, AUROC = 0.816; test: 72.5 {\%}, AUROC = 0.823). Entropy features were significantly different between lobular and ductal cancers (p < 0.001; Mann-Whitney U). IHC classifications using COM features were also similar for training and test data (training: 57.2 {\%}, AUROC = 0.754; test: 57.0 {\%}, AUROC = 0.750). Hormone receptor positive and negative cancers demonstrated significantly different entropy features. Entropy features alone were unable to create a robust classification model. Conclusion: Textural differences on contrast-enhanced MR images may reflect underlying lesion subtypes, which merits testing against treatment response. ",
    keywords = "Breast cancer, Classification, Histological subtypes and immunohistochemical profiles, Magnetic Resonance Imaging (MRI), Texture analysis (TA)",
    author = "Waugh, {S. A.} and Purdie, {C. A.} and Jordan, {L. B.} and S. Vinnicombe and Lerski, {R. A.} and P. Martin and Thompson, {A. M.}",
    year = "2016",
    month = "2",
    doi = "10.1007/s00330-015-3845-6",
    language = "English",
    volume = "26",
    pages = "322--330",
    journal = "European Radiology",
    issn = "0938-7994",
    publisher = "Springer Verlag",
    number = "2",

    }

    Waugh, SA, Purdie, CA, Jordan, LB, Vinnicombe, S, Lerski, RA, Martin, P & Thompson, AM 2016, 'Magnetic resonance imaging texture analysis classification of primary breast cancer', European Radiology, vol. 26, no. 2, pp. 322-330. https://doi.org/10.1007/s00330-015-3845-6

    Magnetic resonance imaging texture analysis classification of primary breast cancer. / Waugh, S. A. (Lead / Corresponding author); Purdie, C. A.; Jordan, L. B.; Vinnicombe, S.; Lerski, R. A.; Martin, P.; Thompson, A. M.

    In: European Radiology, Vol. 26, No. 2, 02.2016, p. 322-330.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Magnetic resonance imaging texture analysis classification of primary breast cancer

    AU - Waugh, S. A.

    AU - Purdie, C. A.

    AU - Jordan, L. B.

    AU - Vinnicombe, S.

    AU - Lerski, R. A.

    AU - Martin, P.

    AU - Thompson, A. M.

    PY - 2016/2

    Y1 - 2016/2

    N2 - Objectives: Patient-tailored treatments for breast cancer are based on histological and immunohistochemical (IHC) subtypes. Magnetic Resonance Imaging (MRI) texture analysis (TA) may be useful in non-invasive lesion subtype classification. Methods: Women with newly diagnosed primary breast cancer underwent pre-treatment dynamic contrast-enhanced breast MRI. TA was performed using co-occurrence matrix (COM) features, by creating a model on retrospective training data, then prospectively applying to a test set. Analyses were blinded to breast pathology. Subtype classifications were performed using a cross-validated k-nearest-neighbour (k = 3) technique, with accuracy relative to pathology assessed and receiver operator curve (AUROC) calculated. Mann-Whitney U and Kruskal-Wallis tests were used to assess raw entropy feature values. Results: Histological subtype classifications were similar across training (n = 148 cancers) and test sets (n = 73 lesions) using all COM features (training: 75 %, AUROC = 0.816; test: 72.5 %, AUROC = 0.823). Entropy features were significantly different between lobular and ductal cancers (p < 0.001; Mann-Whitney U). IHC classifications using COM features were also similar for training and test data (training: 57.2 %, AUROC = 0.754; test: 57.0 %, AUROC = 0.750). Hormone receptor positive and negative cancers demonstrated significantly different entropy features. Entropy features alone were unable to create a robust classification model. Conclusion: Textural differences on contrast-enhanced MR images may reflect underlying lesion subtypes, which merits testing against treatment response. 

    AB - Objectives: Patient-tailored treatments for breast cancer are based on histological and immunohistochemical (IHC) subtypes. Magnetic Resonance Imaging (MRI) texture analysis (TA) may be useful in non-invasive lesion subtype classification. Methods: Women with newly diagnosed primary breast cancer underwent pre-treatment dynamic contrast-enhanced breast MRI. TA was performed using co-occurrence matrix (COM) features, by creating a model on retrospective training data, then prospectively applying to a test set. Analyses were blinded to breast pathology. Subtype classifications were performed using a cross-validated k-nearest-neighbour (k = 3) technique, with accuracy relative to pathology assessed and receiver operator curve (AUROC) calculated. Mann-Whitney U and Kruskal-Wallis tests were used to assess raw entropy feature values. Results: Histological subtype classifications were similar across training (n = 148 cancers) and test sets (n = 73 lesions) using all COM features (training: 75 %, AUROC = 0.816; test: 72.5 %, AUROC = 0.823). Entropy features were significantly different between lobular and ductal cancers (p < 0.001; Mann-Whitney U). IHC classifications using COM features were also similar for training and test data (training: 57.2 %, AUROC = 0.754; test: 57.0 %, AUROC = 0.750). Hormone receptor positive and negative cancers demonstrated significantly different entropy features. Entropy features alone were unable to create a robust classification model. Conclusion: Textural differences on contrast-enhanced MR images may reflect underlying lesion subtypes, which merits testing against treatment response. 

    KW - Breast cancer

    KW - Classification

    KW - Histological subtypes and immunohistochemical profiles

    KW - Magnetic Resonance Imaging (MRI)

    KW - Texture analysis (TA)

    UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-84954382957&origin=resultslist&sort=plf-f&src=s&st1=Magnetic+resonance+imaging+texture+analysis+classification+of+primary+breast+cancer&st2=&sid=416CBE5A16E7062E8C7C58EA9AD33A0F.kqQeWtawXauCyC8ghhRGJg%3a20&sot=b&sdt=b&sl=98&s=TITLE-ABS-KEY%28Magnetic+resonance+imaging+texture+analysis+classification+of+primary+breast+cancer%29&relpos=0&citeCnt=1&searchTerm=

    U2 - 10.1007/s00330-015-3845-6

    DO - 10.1007/s00330-015-3845-6

    M3 - Article

    VL - 26

    SP - 322

    EP - 330

    JO - European Radiology

    JF - European Radiology

    SN - 0938-7994

    IS - 2

    ER -

    Waugh SA, Purdie CA, Jordan LB, Vinnicombe S, Lerski RA, Martin P et al. Magnetic resonance imaging texture analysis classification of primary breast cancer. European Radiology. 2016 Feb;26(2):322-330. https://doi.org/10.1007/s00330-015-3845-6