This is the implementation of the MVCM model mentioned in our paper 'Validation of artificial intelligence contrast mammography in diagnosis of breast cancer: Relationship to histopathological results'.
Paper link here.
We built a multiview deep learning model (MVCM) to classify and segment malignancy in contrast-enhanced mammography images. The model was trained on the CDD-CESM Dataset.
The project was tested on a virtual environment of python 3.7, pip 24.0, and MacOS
- pip install -r full_requirements.txt (or pip install -r requirements.txt if there are errors because of using a different operating system, as requirements.txt only contains the main dependencies and pip will fetch the compatible sub-dependencies, but it will be slower)
- modify configs.py to point to the training/testing sets & control the training flow
- python train.py
- python test.py (to evaluate the model)
- CDD-CESM Dataset here.
To cite this paper, please use:
@article{HELAL2024111392,
title = {Validation of artificial intelligence contrast mammography in diagnosis of breast cancer: Relationship to histopathological results},
journal = {European Journal of Radiology},
volume = {173},
pages = {111392},
year = {2024},
issn = {0720-048X},
doi = {https://doi.org/10.1016/j.ejrad.2024.111392},
url = {https://www.sciencedirect.com/science/article/pii/S0720048X24001086},
author = {Maha Helal and Rana Khaled and Omar Alfarghaly and Omnia Mokhtar and Abeer Elkorany and Aly Fahmy and Hebatalla {El Kassas}}
}