Deep Learning-Based Cataract Detection System
Description
This project aims to develop a deep learning-based mobile app for automated cataract detection in eye images. The system leverages advanced machine learning algorithms And CNN models to accurately identify the presence and severity of cataracts, enabling early diagnosis and treatment.
Project structure
- data/: Directory to store raw and processed datasets, as well as trained models.
- utils/: Utility files
- mobile_app/: Directory containing the mobile application code.
- android/: Android-specific configuration and resources.
- ios/: iOS-specific configuration and resources.
- docs/: Documentation files including project report, setup guides, etc.
- notebooks/: Jupyter notebooks for data analysis and model training.
- README.md: Main project README file containing project overview, setup instructions, etc.
Installation
- Clone the repository:
git clone https://github.com/GeekyYouthsInfo/Deep-Learning-Cataract-Detection.git
- Follow additional setup instructions in the
README.md
file.
Usage
- Follow on-screen instructions to upload eye images and view the detection results.
- For detailed usage guidelines, refer to the documentation.
Model Deployment (Cloud)
- The trained model is deployed on mobile and a cloud platform for scalability and accessibility.
- Access the deployed model through the provided user facing interfaces.
- Submit requests with eye images for cataract detection.
- Receive the detection results through the interface.
Dependencies
- Python 3.7 or higher
- Keras API with Tensor Flow backend
- OpenCV
- Other dependencies listed in
requirements.txt
Tools
I found this tool useful for arugmentation
image augumentor
MOdel Project Architecture
![ResNet hungingface](https://huggingface.co/microsoft/resnet-50)
Contributing
This project is a collaborative effort of a team of four individuals:
- TODO: Led the development of the deep learning models and cloud deployment.
- TODO : Led data preprocessing and algorithm optimization.
- TODO : Led Design of the user interface and conducted usability testing.
- TODO : led documentation and project management tasks.
If you wish to contribute to the project, please follow these steps:
- Fork the repository and create a new branch.
- Make your changes and submit a pull request.
- Ensure that your contributions align with the project's goals and coding standards.
License
This project is licensed under the MIT License. See the LICENSE file for details.
Credits
- DERU HAWULAH NAKATO (Member 1)
- DE GUZMAN ERNESTO ZZIWA (Member 2)
- MASUBA ABDQADIR (Member 3)
- THEMBO JONATHAN (Member 4)
- Special thanks to the TensorFlow and OpenCV communities.
Contact
For questions or inquiries, please contact the project team at https://sites.google.com/view/open-mrs-ecosystem/home