This project focuses on the segmentation of patches of histology images of breast cancer. The goal is to develop and implement a machine learning model that can accurately segment and identify different tissue regions in histology images. This project can be beneficial for educational purposes.
notebooks/: Contains the Jupyter Notebook (breast_cancer_keras.ipynb) with the entire workflow for image segmentation.README.md: Overview of the project and instructions for setting up and running the code.
- Python 3.7+
- Jupyter Notebook
- TensorFlow
- Keras
- NumPy
- OpenCV
- Scikit-learn
- Matplotlib
- Pandas
The dataset used in this project consists of histology images of breast cancer. You may find the dataset at https://pubmed.ncbi.nlm.nih.gov/27563488/
The notebook includes the following steps:
- Data Loading: Load the histology images from the
data/directory. - Data Preprocessing: Preprocess the images, including resizing, normalization, and data augmentation.
- Model Building: Build a segmentation model using a neural network architecture suitable for image segmentation.
- Model Training: Train the model on the preprocessed images.
- Model Evaluation: Evaluate the model's performance using appropriate metrics.
If you would like to contribute to this project, please fork the repository and submit a pull request.
This project is licensed under the MIT License. See the LICENSE file for more details.
For any questions or issues, please open an issue on GitHub or contact me at cesar0407p@gmail.com
Happy coding!