https://github.com/yukta026/eye_disease_prediction
Detection and Classification of Eye Diseases: Diabetic Retinopathy, Cataract, and Glaucoma Using CNN
https://github.com/yukta026/eye_disease_prediction
classification-model cnn-classification cnn-keras cnn-tensorflow data-augmentation machine-learning prediction-model python3
Last synced: 4 months ago
JSON representation
Detection and Classification of Eye Diseases: Diabetic Retinopathy, Cataract, and Glaucoma Using CNN
- Host: GitHub
- URL: https://github.com/yukta026/eye_disease_prediction
- Owner: Yukta026
- Created: 2024-09-08T20:09:28.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-09-10T13:13:35.000Z (9 months ago)
- Last Synced: 2025-01-08T06:36:28.900Z (6 months ago)
- Topics: classification-model, cnn-classification, cnn-keras, cnn-tensorflow, data-augmentation, machine-learning, prediction-model, python3
- Language: Jupyter Notebook
- Homepage:
- Size: 3.23 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Eye Disease Detection Using CNN
## Project Overview
This project aims to **identify and classify eye diseases** from provided eye images, focusing on conditions such as **diabetic retinopathy, cataract, and glaucoma**.
By leveraging deep learning techniques, the project utilizes **Convolutional Neural Networks (CNNs)** to detect these diseases with high accuracy.
## Key Features
- **Deep Learning Framework**: Built using **TensorFlow** and **Keras** libraries.
- **Image Processing**: Eye images are preprocessed, including resizing, rescaling, and data augmentation, to improve model performance.
- **CNN Model**: A robust CNN architecture has been trained to classify eye diseases.
- **Confidence Scores**: For each prediction, the model provides a confidence score, indicating the certainty of its classification.
## Model Highlights
- **Dataset**: The model is trained and validated using a dataset containing labeled images of different eye conditions.
- **Training Strategy**: The data is split into training (80%), validation (10%), and testing (10%) sets to ensure effective model learning and evaluation.
- **Real-time Prediction**: The trained model can be used to make predictions on new, unseen images, outputting both the predicted disease and the confidence level of that prediction.
## Applications
This project has significant implications for early detection of eye diseases, aiding in timely diagnosis and treatment, especially in areas where medical resources are limited.
## Demonstration -
## References -
1) https://www.kaggle.com/datasets/gunavenkatdoddi/eye-diseases-classification
2) https://www.youtube.com/watch?v=dGtDTjYs3xc&list=PLeo1K3hjS3ut49PskOfLnE6WUoOp_2lsD