Chest X-Ray Classification with ResNet-50

This project implements a deep learning model for classifying chest X-ray images into three categories: COVID-19, Normal, and Viral Pneumonia. It utilizes a ResNet-50 model with transfer learning, leveraging pre-trained ImageNet weights for improved accuracy.

Dataset

The model is trained and evaluated on the "Covid19-dataset" available on Kaggle: Covid19-dataset

The dataset contains chest X-ray images in JPEG format, divided into 'train' and 'test' folders, with each folder containing subfolders for the three classes:

• Covid
• Normal
• Viral Pneumonia

Model

The model used in this project is ResNet-50, a deep convolutional neural network known for its strong performance in image classification tasks. Transfer learning is employed by using the ResNet-50 model pre-trained on the ImageNet dataset. The final fully connected layer of the ResNet-50 model is replaced with a new layer to perform 3-class classification specific to this project.

Pretrained Model

This project uses a pretrained ResNet-50 model, which is downloaded directly from Kaggle using kagglehub. The model can be found at the following link: ResNet-50 on Kaggle

Requirements

This notebook is intended to be run on Google Colab. The following libraries are used:

• TensorFlow
• Keras
• Scikit-learn
• Matplotlib
• PIL
• Numpy
• Pathlib
• Shutil
• Kagglehub
• OS

Usage

1. Open the notebook in Google Colab: Open in Colab
2. Ensure the runtime is set to GPU (Runtime -> Change runtime type -> Hardware accelerator -> GPU).
3. Run all cells in the notebook sequentially.

License

This project is licensed under the MIT License - see the LICENSE file for details.