---

Breast Cancer Image Segmentation and Classification

![Project Logo/Icon]

Overview

This project aims to facilitate the segmentation and classification of breast cancer ultrasound images. It began as an extension of code initially developed by Khaled Hushme on Kaggle. Our project includes a graphical user interface (GUI) for image segmentation, Gabor feature extraction, and a Convolutional Neural Network (CNN) for image classification. Please note that the classification component is currently a work in progress and is actively being improved.

Features

• Ultrasound Image Segmentation: The GUI enables users to upload and process ultrasound images, perform segmentation to isolate regions of interest (ROIs), and generate masks for further analysis.

• Gabor Feature Extraction: We have incorporated Gabor feature extraction to capture texture and structural information from the segmented ultrasound images. These features can be used for subsequent classification tasks.

• Image Classification: While the classification model's performance is currently being enhanced, it serves as a foundational framework for future improvements and research endeavors.

Getting Started

1. Clone the Repository: Start by cloning this repository to your local machine.

   git clone https://github.com/Isaac-Zimba-J/Breast-cancer-classification.git

2. Dependencies: Ensure you have all the necessary dependencies installed. This includes Python, OpenCV, TensorFlow, and other relevant packages. You can find a list of dependencies in the project's documentation.

3. Usage: Execute the GUI application to segment images, create masks, and apply Gabor feature extraction. These features can then be used in the classification process using the provided CNN model (Note: Improvements to the classification model are ongoing).

4. Contributions: If you are interested in contributing to this project, please refer to our Contribution Guidelines.

Project Structure

• /src: This directory contains the source code for the GUI application, image segmentation, feature extraction, and classification components.

• /data: Store your ultrasound images and datasets in this directory.

• /models: Save pre-trained or trained CNN models here.

• /results: This directory is intended for storing segmentation masks, feature vectors, and classification results.

Future Enhancements

• Enhance the performance of the image classification model by refining the CNN architecture and optimizing hyperparameters.

• Improve the user interface (GUI) for a more user-friendly experience.

• Expand documentation, provide usage examples, and offer tutorials to assist users and contributors.

Contact Information

• GitHub Profile: Isaac-Zimba-J

• Facebook: Zimbapythondev

• WhatsApp: +260970187487

• Email: zimbaisaacj2002@gmail.com

For questions, issues, or potential collaborations, feel free to get in touch via the provided contact information.

---