Chest X-ray Classification System using DenseNet121 in PyTorch Overview This repository contains the implementation of a robust Chest X-ray classification system using the DenseNet121 architecture in PyTorch. The model has achieved an impressive 95% test accuracy through effective pattern recognition and feature extraction. The system demonstrates high precision in image classification, making it a reliable tool for identifying and classifying medical conditions in chest X-ray images.
- DenseNet121 Architecture: The model is built on the DenseNet121 architecture, known for its effectiveness in capturing intricate medical image patterns.
- High Test Accuracy: Achieved a remarkable 95% test accuracy, showcasing the model's proficiency in accurately classifying chest X-ray images.
- Data Augmentation: Addressed data imbalance challenges through strategic data augmentation techniques. This enhancement improves the model's ability to generalize across COVID-19, Pneumonia, and Normal classes, resulting in improved overall classification performance.
Visit our online demo to experience the Chest X-ray Classification in action:
Chest X-ray Classification Demo
For users interested in fine-tuning the model or using it for their specific applications, the pre-trained model state dict can be downloaded from the following Hugging Face model hub link:
Download Deployment Files and DenseNet121 Model State Dict
The architecture is based on DenseNet121, a deep convolutional neural network known for its effectiveness in capturing intricate patterns in images.