Public demo of MammaScope, a Streamlit-based software tool for integrating and comparing immunohistochemistry (IHC) and MammaTyper molecular profiling in breast cancer.
The following diagram summarizes the main processing pipeline of the demo application, from file upload to result generation.
flowchart LR
A[User / Researcher] --> B[Streamlit Demo Interface]
B --> C1[Upload IHC Excel file<br/>PatWin data]
B --> C2[Upload MammaTyper PDF report]
C1 --> D1[Validate and preprocess<br/>IHC data]
C2 --> D2[Parse PDF and extract<br/>molecular biomarkers]
D1 --> E[Integrated data model]
D2 --> E
E --> F[Subtype comparison engine]
F --> G1[Concordant cases]
F --> G2[Discordant cases]
G1 --> H[Structured results]
G2 --> H
H --> I[Interactive visualization<br/>in Streamlit]
H --> J[Exportable outputs<br/>reports / tables]
Try the application online:
https://mammascope-demo.streamlit.app/
The demo includes simulated and anonymized example files so the full workflow can be tested without using real clinical data.
Breast cancer molecular subtyping is essential for guiding therapeutic decisions. In routine clinical practice, classification is commonly performed using immunohistochemistry (IHC) by evaluating biomarkers such as:
- ER (Estrogen Receptor)
- PR (Progesterone Receptor)
- HER2
- Ki-67
However, IHC presents limitations related to inter-observer variability and subjective interpretation, particularly for Ki-67.
The MammaTyper® assay, based on RT-qPCR technology, quantifies the expression of the genes:
- ESR1
- PGR
- ERBB2
- MKI67
This provides a quantitative and reproducible molecular classification.
This project develops a software tool that automates the integration and comparison of results between both methods, facilitating the analysis of diagnostic concordance.
The system provides the following functionality:
- Import of IHC results from Excel files (PatWin)
- Import of MammaTyper PDF reports
- Automatic biomarker extraction
- Integration of results into a structured dataset
- Identification of concordances and discordances
- Automatic report generation
- Export of processed results
The application is implemented using Python and Streamlit, providing a lightweight interface suitable for clinical environments.
Clone the repository:
git clone https://github.com/diegoalvrezz/MammaScope-Demo.git
cd MammaScope-DemoInstall dependencies:
pip install -r requirements.txtRun the application:
streamlit run demo_app/demo_app.pyThe application will open automatically in your browser.
Example anonymized files are included in:
demo_app/demo_files
These files allow users to test the complete workflow without requiring real hospital data.
MammaScope-Demo
│
├── codigo/ core processing modules
│
├── demo_app/ Streamlit demo application
│ ├── demo_app.py
│ ├── ajustes.py
│ ├── extraccion.py
│ ├── informes.py
│ ├── discordancia.py
│ ├── db.py
│ ├── auth.py
│ ├── vista_historico.py
│ ├── stats_biomarcadores.py
│ └── demo_files/
│
├── requirements.txt
├── README.md
└── LICENSE
This repository does not contain real clinical data.
All files included in the demo are:
- simulated
- anonymized
- intended only for demonstration purposes
The full application is designed to operate with previously anonymized data within a hospital environment.
Diego Vallina Álvarez
Health Engineering Degree
University of Burgos
Bachelor’s Thesis developed in collaboration with the Hospital Universitario de Burgos (HUBU).
This project is distributed under the MIT License.
See the LICENSE file for details.