This project presents a cost-efficient and accurate method for detecting and classifying retinal diseases from OCT images using advanced Convolutional Neural Networks (CNNs) and Transformer-based architectures. Developed with a focus on precision and scalability, it aims to aid early diagnosis and prevention of retinal conditions, ensuring enhanced patient care.
- Hybrid Architecture: Combines CNN for feature extraction and Transformer for contextual analysis.
- Optimized Workflow: Implements image segmentation for better feature extraction, reducing processing time.
- High Accuracy: Achieves superior results in retinal disease detection and classification.
- Prevention Insights: Provides actionable insights to prevent disease progression.
- Retinal Diseases Detection: Code and data for training and testing the model.
- Result Pictures: Visual outputs showcasing classification results.
- README.md: Documentation and project insights.
The model demonstrates impressive performance in detecting and classifying various retinal conditions, significantly contributing to automated diagnostics.
- Languages: Python
- Libraries: TensorFlow, OpenCV, NumPy, Matplotlib
- Techniques: Deep Learning, Image Processing, Transformers, Image Classification
- Clone the repository: git clone https://github.com/Zuboy/Retinal-Disease-Classficiation-using-DL.git
- Navigate to the project directory and install dependencies: pip install -r requirements.txt
- Execute the main script to train or test the model: python main.py
This project was created as part of a Bachelor's program at JSPM Jayawantrao Sawant College of Engineering (2023–2024) to explore the potential of deep learning in healthcare.
Contributions are welcome! Feel free to fork the repository and submit pull requests for improvements.