Segmentation of the Main Structures in Hematoxylin and Eosin Images

Pathologists conduct a biopsy on a tissue when a carcinoma case for a patient is suspected. They stain the cells on that tissue using some biochemical materials that react with a certain cell element. They put stained cells onto a slide and examine the cells by using an optical microscope device. In our case, we will focus on H&E stained breast tissue samples. Pathologists keep track of a standard process to determine the patient’s condition by focusing on the structures in H&E stained images such as epithelium, lumen, and nuclei. They employ scoring methods with quantitative and qualitative inferences in this decision process. Those inferences contain mitotic nuclei activity, number of the nucleus, lumen region distribution, epithelium area size and so on. Each factor has a score for the patient’s carcinoma case. In this paper a novel image processing algorithm is developed to enable the pathologists to make decisions easily by segmenting epithelium, lumen and nuclei structures. Actual microscopic images could show some degenerated cell structures because of staining variability and some artifacts. Our algorithm demonstrates the structures clearly while colorizing them with distinctive colors considering their transparency.

Download Full Text

Hematoxylin-Eosin staining, Digital pathology, Image processing, SLIC

Quantitative output variables

In this section, the evaluation process of the proposed algorithm is provided. This process includes gathering ground truth data and the outputs of our algorithm on that data. The histology slides were scanned on a Hamamatsu NanoZoomer C9600. These histology samples were obtained from breast tissue. Figure 4 shows two input images and their corresponding results. According to those results, lumen regions are colored with green, epithelium regions are colored in red and nuclei are colored in blue. As a consequence, expressing the real borders of the main structures in H&E images enable pathologist to examine the cells easily. Colorized structures by the proposed algorithm seem more remarkable and distinguishable for doctors.


[1] C. W. Elston and I. O. Ellis, "Pathological prognostic factors in breast cancer. i. the value of histological grade in breast cancer: experience from a large study with long-term follow-up," Histopathology, vol. 19, no. 5, pp. 403–410, 1991.
[2] V. Roullier, O. Lezoray, V.-T. Ta, and A. Elmoataz, "Multi-resolution ´ graph-based analysis of histopathological whole slide images: Application to mitotic cell extraction and visualization," Computerized Medical Imaging and Graphics, vol. 35, no. 7-8, pp. 603–615, 2011.
[3] M. D. Zarella, C. Yeoh, D. E. Breen, and F. U. Garcia, "An alternative reference space for h&e color normalization," PloS one, vol. 12, no. 3, p. e0174489, 2017
[4] M. Veta, P. J. van Diest, R. Kornegoor, A. Huisman, M. A. Viergever, and J. P. Pluim, "Automatic nuclei segmentation in h&e stained breast cancer histopathology images," PloS one, vol. 8, no. 7, p. e70221, 2013.
[5] R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua, and S. Susstrunk, ¨ "Slic superpixels compared to state-of-the-art superpixel methods," IEEE transactions on pattern analysis and machine intelligence, vol. 34, no. 11, pp. 2274–2282, 2012.
[6] S. Suzuki et al., "Topological structural analysis of digitized binary images by border following," Computer vision, graphics, and image processing, vol. 30, no. 1, pp. 32–46, 1985.
[7] J. N. Kapur, P. K. Sahoo, and A. K. Wong, "A new method for graylevel picture thresholding using the entropy of the histogram," Computer vision, graphics, and image processing, vol. 29, no. 3, pp. 273–285, 1985.
[8] G. Loy and A. Zelinsky, "Fast radial symmetry for detecting points of interest," IEEE Transactions on pattern analysis and machine intelligence, vol. 25, no. 8, pp. 959–973, 2003.
[9] R. Barnes, C. Lehman, and D. Mulla, "Priority-flood: An optimal depression-filling and watershed-labeling algorithm for digital elevation models," Computers & Geosciences, vol. 62, pp. 117–127, 2014.

Other Articles

An Automated and Accurate Methodology to Assess Ki-67 Labeling Index of Immunohistochemical Staining Images of Breast Cancer Tissues

Automatic scoring of Ki-67 with digital image analysis would improve the accuracy of the diagnostic.

Read More

Automated Cell Segmentation and Spot Detection in Fluorescence in Situ Hybridization Staining to Assess HER2 Status in Breast Cancer

Fluorescence in situ hybridization (FISH) approach is constituted of a pair of complementary techniques for precisely detecting...

Read More

Contact us
for more information

For more information about our solutions, you can contact us and request a demo.