Back to Showcase
The Fundus Cataract Detection System is a web-based application developed using Python and Flask to classify fundus retinal images as Cataract or Normal using a deep learning model. The system is designed to assist in early cataract screening by automatically analyzing uploaded fundus images and providing classification results with high accuracy.
Before deployment, the deep learning model is trained offline using a dataset of more than 10,000 labeled fundus images for training and validation. The resulting trained model is then integrated into the Flask application as the inference engine, allowing the system to perform real-time image classification without retraining.
The application supports a single user role (Administrator). Administrators can register user accounts, while each registered user has access only to their own uploaded images and classification history, ensuring data privacy and organized record management.
Project Summary
Before deployment, the deep learning model is trained offline using a dataset of more than 10,000 labeled fundus images for training and validation. The resulting trained model is then integrated into the Flask application as the inference engine, allowing the system to perform real-time image classification without retraining.
The application supports a single user role (Administrator). Administrators can register user accounts, while each registered user has access only to their own uploaded images and classification history, ensuring data privacy and organized record management.
System Galleries