Identification of immune cell infiltration in hematoxylin-eosin stained breast cancer samples: texture-based classification of tissue morphologies

The characteristics of immune cells in the tumor microenvironment of breast cancer capture clinically important information. Despite the heterogeneity of tumor-infiltrating immune cells, it has been shown that the degree of infiltration assessed by visual evaluation of hematoxylin-eosin (H and E) stained samples has prognostic and possibly predictive value. However, quantification of the infiltration in H and E-stained tissue samples is currently dependent on visual scoring by an expert. Computer vision enables automated characterization of the components of the tumor microenvironment, and texture-based methods have successfully been used to discriminate between different tissue morphologies and cell phenotypes. In this study, we evaluate whether local binary pattern texture features with superpixel segmentation and classification with support vector machine can be utilized to identify immune cell infiltration in H and E-stained breast cancer samples. Guided with the pan-leukocyte CD45 marker, we annotated training and test sets from 20 primary breast cancer samples. In the training set of arbitrary sized image regions (n=1,116) a 3-fold cross-validation resulted in 98% accuracy and an area under the receiver-operating characteristic curve (AUC) of 0.98 to discriminate between immune cell -rich and - poor areas. In the test set (n=204), we achieved an accuracy of 96% and AUC of 0.99 to label cropped tissue regions correctly into immune cell -rich and -poor categories. The obtained results demonstrate strong discrimination between immune cell -rich and -poor tissue morphologies. The proposed method can provide a quantitative measurement of the degree of immune cell infiltration and applied to digitally scanned H and E-stained breast cancer samples for diagnostic purposes.

[1]  T. Nielsen,et al.  The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014. , 2015, Annals of oncology : official journal of the European Society for Medical Oncology.

[2]  S Michiels,et al.  Tumor infiltrating lymphocytes are prognostic in triple negative breast cancer and predictive for trastuzumab benefit in early breast cancer: results from the FinHER trial. , 2014, Annals of oncology : official journal of the European Society for Medical Oncology.

[3]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  F. Markowetz,et al.  Quantitative Image Analysis of Cellular Heterogeneity in Breast Tumors Complements Genomic Profiling , 2012, Science Translational Medicine.

[5]  Georgios Tziritas,et al.  Lymphocyte segmentation using the transferable belief model , 2010, ICPR 2010.

[6]  Anthony G Gallagher,et al.  Do we see what we think we see? The complexities of morphological assessment , 2009, The Journal of pathology.

[7]  Gyan Bhanot,et al.  Computerized Image-Based Detection and Grading of Lymphocytic Infiltration in HER2+ Breast Cancer Histopathology , 2010, IEEE Transactions on Biomedical Engineering.

[8]  Chih-Jen Lin,et al.  LIBLINEAR: A Library for Large Linear Classification , 2008, J. Mach. Learn. Res..

[9]  Matti Pietikäinen,et al.  Identification of tumor epithelium and stroma in tissue microarrays using texture analysis , 2012, Diagnostic Pathology.

[10]  Max A. Viergever,et al.  Breast Cancer Histopathology Image Analysis: A Review , 2014, IEEE Transactions on Biomedical Engineering.

[11]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  J. Elmore,et al.  Diagnostic concordance among pathologists interpreting breast biopsy specimens. , 2015, JAMA.

[13]  Andrew Zisserman,et al.  Sparse kernel approximations for efficient classification and detection , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[14]  Gyan Bhanot,et al.  Expectation–Maximization-Driven Geodesic Active Contour With Overlap Resolution (EMaGACOR): Application to Lymphocyte Segmentation on Breast Cancer Histopathology , 2010, IEEE Transactions on Biomedical Engineering.

[15]  Manohar Kuse,et al.  A Classification Scheme for Lymphocyte Segmentation in H&E Stained Histology Images , 2010, ICPR Contests.