Histological Detection of High-Risk Benign Breast Lesions from Whole Slide Images

  • Akif Burak TosunEmail author
  • Luong NguyenEmail author
  • Nathan Ong
  • Olga Navolotskaia
  • Gloria Carter
  • Jeffrey L. Fine
  • D. Lansing Taylor
  • S. Chakra Chennubhotla
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10434)


Accurate diagnosis of high-risk benign breast lesions is crucial in patient management since they are associated with an increased risk of invasive breast cancer development. Since it is not yet possible to identify the occult cancer patients without surgery, this limitation leads to retrospectively unnecessary surgeries. In this paper, we present a computational pathology pipeline for histological diagnosis of high-risk benign breast lesions from whole slide images (WSIs). Our pipeline includes WSI stain color normalization, ductal regions of interest (ROIs) segmentation, and cytological and architectural feature extraction to classify ductal ROIs into triaged high-risk benign lesions. We curated 93 WSIs of breast tissues containing high-risk benign lesions based on pathology reports and collected ground truth annotations from three different pathologists for the ductal ROIs segmented by our pipeline. Our method has comparable performance to a pool of expert pathologists.


Breast lesions Atypical ductal hyperplasia Computational pathology Pattern recognition Architectural pattern Classification 


  1. 1.
    Bejnordi, B., et al.: Automated detection of DCIS in whole-slide H&E stained breast histopathology images. IEEE-TMI 35(9), 2141–2150 (2016)Google Scholar
  2. 2.
    Calhoun, B., et al.: Management of flat epithelial atypia on breast core biopsy may be individualized based on correlation with imaging studies. Mod. Pathol. 28(5), 670–676 (2015)CrossRefGoogle Scholar
  3. 3.
    Dong, F., et al.: Computational pathology to discriminate benign from malignant intraductal proliferations of the breast. PLoS One 9(12), e114885 (2014)CrossRefGoogle Scholar
  4. 4.
    Dundar, M., et al.: Computerized classification of intraductal breast lesions using histopathological images. IEEE-TBE 58(7), 1977–1984 (2011)Google Scholar
  5. 5.
    Dupont, W., Page, D.: Risk factors for breast cancer in women with proliferative breast disease. N. Engl. J. Med. 312(3), 146–151 (1985)CrossRefGoogle Scholar
  6. 6.
    Elmore, J., et al.: Diagnostic concordance among pathologists interpreting breast biopsy specimens. JAMA 313(11), 1122–1132 (2015)CrossRefGoogle Scholar
  7. 7.
    Khan, A., et al.: A nonlinear mapping approach to stain normalization in digital histopathology images using image-specific color deconvolution. IEEE-TBE 61(6), 1729–1738 (2014)Google Scholar
  8. 8.
    Krizhevsky, et al.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)Google Scholar
  9. 9.
    Nguyen, L., et al.: Architectural patterns for differential diagnosis of proliferative breast lesions from histopathological images. In: IEEE-ISBI (2017)Google Scholar
  10. 10.
    Nguyen, L., et al.: Spatial statistics for segmenting histological structures in H&E stained tissue images. IEEE-TMI PP(99), 1 (2017)Google Scholar
  11. 11.
    Pinder, S., Reis-Filho, J.: Non-operative breast pathology: columnar cell lesions. J. Clin. Pathol. 60(12), 1307–1312 (2007)CrossRefGoogle Scholar
  12. 12.
    Said, S., et al.: Flat epithelial atypia and risk of breast cancer: a mayo cohort study. Cancer 121(10), 1548–1555 (2015)CrossRefGoogle Scholar
  13. 13.
    Schindelin, J., et al.: Fiji: an open-source platform for biological-image analysis. Nat. Methods 9(7), 676–682 (2012)CrossRefGoogle Scholar
  14. 14.
    Sermanet, P., et al.: Overfeat: Integrated recognition, localization and detection using convolutional networks. arXiv:1312.6229 (2013)
  15. 15.
    Srinivas, U., et al.: SHIRC: a simultaneous sparsity model for histopathological image representation and classification. In: IEEE-ISBI, pp. 1118–1121 (2013)Google Scholar
  16. 16.
    Tosun, A., Gunduz-Demir, C.: Graph run-length matrices for histopathological image segmentation. IEEE-TMI 30(3), 721–732 (2011)Google Scholar
  17. 17.
    Tosun, A., et al.: Object-oriented texture analysis for the unsupervised segmentation of biopsy images for cancer detection. Pattern Recogn. 42(6), 1104–1112 (2009)CrossRefGoogle Scholar
  18. 18.
    Vahadane, A., et al.: Structure-preserving color normalization and sparse stain separation for histological images. IEEE-TMI 35(8), 1962–1971 (2016)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Akif Burak Tosun
    • 1
    Email author
  • Luong Nguyen
    • 1
    Email author
  • Nathan Ong
    • 1
  • Olga Navolotskaia
    • 2
  • Gloria Carter
    • 2
  • Jeffrey L. Fine
    • 2
  • D. Lansing Taylor
    • 1
    • 3
  • S. Chakra Chennubhotla
    • 1
  1. 1.Department of Computational and Systems BiologyUniversity of PittsburghPittsburghUSA
  2. 2.Department of PathologyMagee Womens Hospital of UPMCPittsburghUSA
  3. 3.Drug Discovery InstituteUniversity of PittsburghPittsburghUSA

Personalised recommendations