Breast cancer tissue stained with PR.
#10172
PR, Breast Cancer, AI
In immunohistochemistry (IHC), progesterone receptor (PR) is an important biomarker in the investigation of breast cancer [1]. This APP is developed for use in discovery of medical knowledge related to PR status in breast cancer.
In order to assess the PR status in breast cancer automatically, robust nuclei detection and segmentation is important. This APP utilizes artificial intelligence (AI) for automatic and robust nuclei detection and segmentation in PR 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 890 manually detected nuclei stained for PR, 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) [2] which captures the segmentation quality. Testing on 890 manually segmented nuclei, the performance measure shows excellent nuclei segmentation as seen from the table below.
| Precision | Recall (Sensitivity) | F1-score | AJI | |
|---|---|---|---|---|
| Mean | 96.5 % | 95.4 % | 0.959 | 0.794 |
After segmentation of the nuclei, the APP is configured to divide them into PR negative, 1+, 2+, or 3+ positive based on the DAB intensity. The user can adjust the threshold for differentiation between the nuclei to fit their research.
The assessment of PR status by the APP has shown good correlation with manual assessment as seen from the figure below. A total of 243 PR breast cancer samples have been used for the comparison.
Outlining of tumor regions in breast cancer tissue stained with PR. 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 “10172 – PR, Breast Cancer, AI” to the outlined tumor regions. Automatic nuclei segmentation is achieved using AI and classification into negative, 1+, 2+, and 3+ is based on the intensity of DAB staining in each nucleus.
Microscopic view of original image
Details
Quantitative Output variables
The APP can output the following variables:
- Total Nuclei (#): The total number of nuclei
- Negative Nuclei (#): The number of negative nuclei
- Positive Nuclei (#): The number of positive nuclei
- Positive Percentage (%): The percentage of positive cells among the total number of cells
- H-Score [0-300]
Workflow
Step 1: Load an ER 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 “10172 – PR, Breast Cancer, AI”
Methods
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 [3].
Keywords
Progesterone Receptor, PR Status, Nuclei Segmentation, Immunohistochemistry, Breast Cancer, Artificial Intelligence, AI, Digital Pathology
References
LITERATURE
1. Hammond, M. E., et al. American Society of Clinical Oncology/College of American Pathologists Guideline Recommendations for Immunohistochemical Testing of Estrogen and Progesterone Receptors in Breast Cancer, Journal of Clinical Oncology, 2010, 28 (16), 2784-95. DOI
2. 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
3. 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
