Empirical comparison of color normalization methods for epithelial-stromal classification in H and E images - PubMed (original) (raw)

Empirical comparison of color normalization methods for epithelial-stromal classification in H and E images

Amit Sethi et al. J Pathol Inform. 2016.

Abstract

Context: Color normalization techniques for histology have not been empirically tested for their utility for computational pathology pipelines.

Aims: We compared two contemporary techniques for achieving a common intermediate goal - epithelial-stromal classification.

Settings and design: Expert-annotated regions of epithelium and stroma were treated as ground truth for comparing classifiers on original and color-normalized images.

Materials and methods: Epithelial and stromal regions were annotated on thirty diverse-appearing H and E stained prostate cancer tissue microarray cores. Corresponding sets of thirty images each were generated using the two color normalization techniques. Color metrics were compared for original and color-normalized images. Separate epithelial-stromal classifiers were trained and compared on test images. Main analyses were conducted using a multiresolution segmentation (MRS) approach; comparative analyses using two other classification approaches (convolutional neural network [CNN], Wndchrm) were also performed.

Statistical analysis: For the main MRS method, which relied on classification of super-pixels, the number of variables used was reduced using backward elimination without compromising accuracy, and test - area under the curves (AUCs) were compared for original and normalized images. For CNN and Wndchrm, pixel classification test-AUCs were compared.

Results: Khan method reduced color saturation while Vahadane reduced hue variance. Super-pixel-level test-AUC for MRS was 0.010-0.025 (95% confidence interval limits ± 0.004) higher for the two normalized image sets compared to the original in the 10-80 variable range. Improvement in pixel classification accuracy was also observed for CNN and Wndchrm for color-normalized images.

Conclusions: Color normalization can give a small incremental benefit when a super-pixel-based classification method is used with features that perform implicit color normalization while the gain is higher for patch-based classification methods for classifying epithelium versus stroma.

Keywords: Color normalization; computational pathology; epithelial-stromal classification.

PubMed Disclaimer

Figures

Figure 1

Figure 1

Diversity of H and E stained images illustrated using four prostate cancer samples with Gleason Grade 3. The first two samples show range of epithelial brightness, and the last two show the range of stromal brightness

Figure 2

Figure 2

Preparation of the training and testing sets using original and color-normalized images

Figure 3

Figure 3

Color normalization illustrated using inter-image standard deviation (error bars) of mean (bars) hue, saturation, and intensity for epithelium and stroma. Continuous color bars between epithelium and stroma illustrate the hue, saturation, and intensity range holding the other two at their means

Figure 4

Figure 4

Contribution of color to epithelial-stromal classification illustrated using mean intra-image standard deviation (error bars) around mean (bars) hue, saturation, and intensity. Color bars between epithelium and stroma illustrate the full range of hue, saturation, and intensity while holding the other two constant

Figure 5

Figure 5

Selected sub-images and their normalized versions

Figure 6

Figure 6

An example H and E stained tissue image (left), its super-pixel boundaries (center, green), and detected dark sub-objects (right, black)

Figure 7

Figure 7

Test - area under the curve for the three sets of images for logistic regression models using different number of variables

Figure 8

Figure 8

Receiver operating characteristic curves for pixel-level accuracy for twenty-feature models for ten test images for the three models for thresholds 0.75, 0.5, 0.25, 0.15, and 0.1 on the logistic regression output

Figure 9

Figure 9

Two examples of cores whose pixels have been classified into epithelium (green) and stroma (red) based on original images as well as normalized images using Vahadane and Khan methods

Similar articles

Cited by

References

    1. Tabesh A, Teverovskiy M, Pang HY, Kumar VP, Verbel D, Kotsianti A, et al. Multifeature prostate cancer diagnosis and Gleason grading of histological images. IEEE Trans Med Imaging. 2007;26:1366–78. - PubMed
    1. Beck AH, Sangoi AR, Leung S, Marinelli RJ, Nielsen TO, van de Vijver MJ, et al. Systematic analysis of breast cancer morphology uncovers stromal features associated with survival. Sci Transl Med. 2011;3:108. - PubMed
    1. Anagnostou VK, Welsh AW, Giltnane JM, Siddiqui S, Liceaga C, Gustavson M, et al. Analytic variability in immunohistochemistry biomarker studies. Cancer Epidemiol Biomarkers Prev. 2010;19:982–91. - PMC - PubMed
    1. Vahadane A, Peng T, Albarqouni S, Baust M, Steiger K, Schlitter AM, et al. Structure-Preserved Color Normalization for Histological Images. International Symposium on Biomedical Imaging (ISBI): From Nano to Macro, New York, USA; April. 2015
    1. Khan AM, Rajpoot N, Treanor D, Magee D. A nonlinear mapping approach to stain normalization in digital histopathology images using image-specific color deconvolution. IEEE Trans Biomed Eng. 2014;61:1729–38. - PubMed

LinkOut - more resources