This project aims to detect lung and colon cancer from histopathological images using a U-Net Convolutional Neural Network (CNN) architecture. The model processes a dataset of 25,000 images, categorized into five different classes, to identify patterns indicative of cancerous tissues. This approach showcases the potential of deep learning in medical image analysis and contributes to early and accurate cancer detection.
Python NumPy TensorFlow scikit-learn Pillow (PIL) OS module pandas datetime random
The dataset used in this project is sourced from Kaggle. It comprises 25,000 histopathological images, divided into five distinct classes. These images provide the basis for training and testing the U-Net model. Link to dataset: https://www.kaggle.com/datasets/andrewmvd/lung-and-colon-cancer-histopathological-images
The core of this project is a U-Net CNN Architecture, known for its efficiency in image segmentation tasks. The model is trained on the provided dataset, learning to differentiate between various cancerous and non-cancerous tissue types.
This project is open-sourced under the MIT License.
For any inquiries or collaboration requests, please contact me at [[email protected]].
Special thanks to the Kaggle community for providing the dataset and to all contributors who have offered valuable insights into the project's development.