Breast cancer tissue stained with Ki-67.
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 . In immunohistochemistry (IHC), Ki-67 is an important biomarker in the investigation of breast cancer [2, 3]. This APP is developed for use in discovery of medical knowledge related to Ki-67 status in breast cancer.
In order to assess the Ki-67 status automatically, robust nuclei detection and segmentation is important. This APP utilizes artificial intelligence (AI) for automatic and robust nuclei detection and segmentation in Ki-67 stained breast cancer tissue.
The nuclei detection performance has been assessed using precision and recall. Precision measures the ability of the APP to only detect viable nuclei and not mistake e.g. artefacts for nuclei. Recall (also known as sensitivity) measures the ability of the APP to detect all nuclei present. Precision and recall are combined to a F1-score which reflects both in a harmonic mean. Testing on 862 manually detected nuclei stained for Ki-67, all three performance measures show excellent nuclei detection as seen from the table below.
The nuclei segmentation performance has been assessed using aggregated Jaccard index (AJI)  which captures the segmentation quality. Testing on 862 manually segmented nuclei, the performance measure shows excellent nuclei segmentation as seen from the table below.
|Mean||96.6 %||96.3 %||0.960||0.805|
After segmentation of the nuclei, the APP is configured to divide them into Ki-67 negative or positive based on the DAB intensity. The user can adjust the threshold for differentiation between the nuclei to fit their research.
The assessment of Ki-67 status by the APP has shown good correlation with manual assessment as seen from the figure below. A total of 181 Ki-67 breast cancer samples have been used for the comparison.
Breast cancer tissue stained with Ki-67.
Outlining of tumor regions in breast cancer tissue stained with Ki-67. Outlining can be done manually, or it can be done automatically using APP “10162 – IHC, Tumor Detection, AI”, as in this case, or using our virtual double staining technique.
Result of applying APP “10173 – Ki-67, Breast Cancer, AI” to the outlined tumor regions. Automatic nuclei segmentation is achieved using AI and classification into negative/positive is based on the intensity of DAB staining in each nucleus.
Ki-67 whole slide image
The result of applying the following APP sequence on the Ki-67 whole slide image: 1) APP “10162 – IHC, Tumor Detection, AI”, 2) APP “10173 – Ki-67, Breast Cancer, AI”, 3) APP “10114 – Hot Spot”. The result shows a heatmap and a hot spot, where the highest proliferation index is found.
Quantitative Output variables
The APP can output the following variables:
Step 1: Load a Ki-67 stained breast cancer image
Step 2: Outline tumor either manually or automatically using one of Visiopharm’s solutions
Step 3: Load and run the APP “10173 – Ki-67, Breast Cancer, AI”
The APP uses a deep learning network, which has been trained using 13,434 annotated nuclei from IHC stained breast cancer tissue, to segment nuclei. The architectural structure of the network is a U-Net which is popular for medical image segmentation. The neural network uses a cascade of layers of nonlinear processing units for feature extraction and transformation, with each successive layer using the output from the previous layers as input. U-Net uses an encoder-decoder structure with a contracting path and an expansive path. For more information on the network architecture, see .
Ki-67, Proliferation Index, Nuclei Segmentation, Immunohistochemistry, Breast Cancer, Artificial Intelligence, AI, Digital Pathology
1. Scholzen, T., et al. The Ki-67 Protein: From the Known and the Unknown, Journal of Cellular Physiology, 2000, 182 (3), 311-22. DOI
2. NordiQC Epitope. Accessed: April 15, 2020.
3. Gnant, M., et al. St. Gallen 2011: Summary of the Consensus Discussion, Breast Care, 2011, 6 (2), 136-141. DOI
4. Kumar, N., et al. A Dataset and a Technique for Generalized Nuclear Segmentation for Computational Pathology, IEEE Transactions on Medical Imaging, 2017, 36 (7), 1550-1560. DOI
5. Ronneberger, O., et al. U-Net: Convolutional Networks for Biomedical Image Segmentation, International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015, 234-241. DOI