After After
Before Before

10005 - Ki-67+MART1, Melanoma


The Ki-67 protein is associated with cellular proliferation, and the protein is present in the nucleus of all cells that are in the active phase of the cell cycle, but absent in resting cells [See REFERENCES: 1]. The cell proliferation rate can be assessed by Ki-67-immunohistochemical (IHC) staining, and this can be correlated to the tumor grade and the clinical course for the patient [See REFERENCES: 2]. For melanocytic lesions, the number and distribution of Ki-67-positive cells can be used to differentiate melanomas from nevi, because the number of positive cells is significantly increased in melanomas compared to nevi [See REFERENCES: 3]. However, prominent lymphatic infiltrate is often present in melanomas and nevi, which may make it difficult to distinguish the two from each other. To make sure that the Ki-67 staining patterns are not misinterpreted, an IHC cocktail detecting both Ki-67 and the melanocytic marker Melanoma Antigen Recognized by T-cells (MART1, sometimes also called MelanA) can be used to distinguish Ki-67-positive melanocytes from Ki-67-positive non-melanocytes [See REFERENCES: 3]. 

This protocol automatically detects and quantifies Ki-67 positive cells within tumor regions. No manual outlining of the tumor is needed, since the tumor region is automatically identified based on the Ki-67/MART1-IHC double staining. Thus, only an overall region of interest (ROI) has to be defined. The protocol provides the number of positive nuclei within the tumor region as well as the ratio compared to the total number of nuclei within the tumor region.


Ki-67, Ki67, MART1, double staining, melanoma, proliferation, quantitative, digital pathology, image analysis.


The first image processing step involves a segmentation of all nuclei. This is done by assigning a label probability to all pixels in the image, resulting in a label probability image. The label probability image is found by an intensity dictionary - a dictionary with small image patches. The intensity dictionary can be coupled to a label dictionary from which the label probability image is obtained. Based on this image, segmentation of nuclei can be done by choosing the most probable label in each pixel [See REFERENCES: 4]. A method for nuclei separation which is based on shape, size and nuclei probability is used, employing a fully automated watershed-based nuclei segmentation technique. The method is an extension of the method of Jung and Kim [See REFERENCES: 5], where an h-minima transform is used before applying the watershed. The nuclei are subsequently divided into positively stained nuclei and negatively stained nuclei. In addition, the cytoplasm surrounding the nuclei is found and segmented into tumor areas and non-tumor areas depending on the staining level of the cytoplasm [See EXAMPLES: figure 2]. Next, nuclei with an even brown reaction surrounded by cytoplasm that has been classified as belonging to a tumor region are identified as positive nuclei. Furthermore, as a post-processing step, nuclei that are too small are removed. The image obtained after post-processing [See EXAMPLES: figure 3] is used to determine the output variables. 

Note on counting: Analysis of full virtual slides takes place in a tile-by-tile fashion. If not handled appropriately, nuclei that are intersecting with neighboring tile boundaries would be counted twice (or more). By using unbiased counting frames [See REFERENCES: 7], this can be avoided [See EXAMPLES: figure 4]. This principle is implemented in the present APP. Depending on the size of nuclei, the application of this principle could make an important difference.


The output variables obtained from this protocol include: 

- NUC POS: The number of positive nuclei profiles in the tumor region

- NUC TOTAL: The total number of nuclei profiles in the tumor region

- POS RATIO (N) = NUC POS / NUC TOTAL: The ratio of number of positive nuclei profiles

- POS RATIO (A) = Area of NUC POS / Area of NUC TOTAL: The area ratio of area of positive nuclei


Auxiliary APPs are used for additional process steps, e.g. finding Region of Interest (ROI).

There are no Auxiliary APPs available.


There is no staining protocol available


The APP utilizes the EngineTM and Viewer software modules, where EngineTM offers an execution platform to expand processing capability and speed of image analysis. Viewer gives a fast review together with several types of image adjustment properties ex. outlining of regions, annotations and direct measures of distance, curve length, radius, etc. 
By adding the AuthorTM module the APP can be customized to fit other purposes. AuthorTMoffers a comprehensive and dedicated set of tools for creating new fit-for-purpose analysis APPs, and no programming experience is required.



1. The Ki-67 protein: from the known and the unknown, T. Scholzen et al., J. Cell Physiol. 2000, 182 (3): 311-22


3. Immunohistochemical Double Stains Against Ki67/MART1 and HMB45/MITF: Promising Diagnostic Tools in Melanocytic Lesions, P. S. Nielsen et al., Am J Dermatopathol. 2011, 33 (4): 361-70.

4. Learning Dictionaries of Discriminative Image Patches, A. L. Dahl et al., British Machine Vision Conference 2011. 

5. Segmenting Clustered Nuclei Using H-minima Transform-Based Marker Extraction and Contour Parameterization, C. Jung et al., IEEE Transactions on Biomedical Engineering 2010, 57(10): 2600-2604.

6. Learning histopathological patterns, A. Kårsnäs, A. L. Dahl, R. Larsen, Journal of Pathology Informatics, 2(2):12, 2011

7. Unbiased Stereology, C.V. Howard & M.G. Reed, QTP Publications

Field of view showing malignant melanoma in an image of skin tissue double stained by IHC for Ki-67 (DAB) and MART1 (Fast Red).
The same field of view as in Figure 1 after automatic segmentation of nuclei and cytoplasm: DAB-stained Ki-67-positive nuclei (green label), hematoxylin-stained Ki-67-negative nuclei (blue label), Fast red-stained MART1-positive cytoplasm (white label), and unstained MART1-negative cytoplasm (turquoise label).
The same field of view as in Figure 1 after specific classification of nuclei surrounded by MART1-positive cytoplasm: Ki-67-positive tumor nuclei (yellow label), and Ki-67-negative tumor nuclei (blue label).
A field of view comparable to that of Figure 1 including an un-biased counting frame in order to ensure that tumor nuclei are counted only once.