camilababo / covid19-x-ray-detection-cnn-model

Explores the use of a simple CNN model for the detection of COVID-19 in X-ray images. Leveraging a dataset from Kaggle, the pipeline includes preprocessing, model architecture design, performance evaluation, and testing on new cases.

COVID-19 X-ray Detection CNN Model

Overview

This project explores the use of TensorFlow to develop a Convolutional Neural Network (CNN) model to detect COVID-19 infection in X-ray images. The model is trained to classify X-ray images into two categories: normal and infected with COVID-19.

Dataset

The dataset used in this project is available on Kaggle and can be accessed here. It includes X-ray images of patients with COVID-19 as well as normal cases. An additional validation dataset was created by taking batches from the training dataset.

File Structure

.
├── README.md
├── requirements.txt
├── cnn-classifier-pipeline.ipynb
├── data
│ ├── test
│ │ ├── covid_test_boston.jpeg
│ │ ├── princeton_covid.jpg
│ │ └── radiopedia_normal.jpg
├── models
│ └── covid19_xray_detection_model.h5
├── logs
│ ├── train
│ │ ├── events.out.tfevents.1708282339.0.v2
│ ├── validation
│ │ ├── events.out.tfevents.1708282366.1.v2
└── .gitignore

Requirements

• Python 3.x
• Numpy
• Matplotlib
• OpenCV
• TensorFlow
• Keras

Usage

To use this project, you can follow the following steps:

1. Clone the repository to your local machine:

   git clone https://github.com/camilababo/Covid19-X-Ray-Detection-CNN-Model

2. Navigate to the project directory:

   cd Covid19-X-Ray-Detection-CNN-Model

3. Install the required dependencies using pip and the provided requirements.txt file:

   pip install -r requirements.txt

4. Open the provided Jupyter Notebook, cnn-classifier-pipeline.ipynb, using Jupyter and run the pipeline.

5. The trained model file is also available in models/covid19_xray_detection_model.h5.