XRAY-IMAGING

Analyzing X-Ray Imaging of potential covid-patient using deep learning

Disclaimer

The dataset was quite small (180 images each). I do not claim any diagonstic performance of the model this is simply an experiment inspired by this paper

Dependencies

• Tensorflow 2.0
• Keras
• Numpy
• CV2

pip install tensorflow or pip install tensorflow-gpu (For the GPU accelerated version)

pip install opencv-python

Datasets

XRAY Images of COVID Patients vs Normal (PA View)

The datasets are from HERE and Kaggle

Example

python3 test.py sample.jpeg models/100accuracy

The command above should print "Negative"

Architecture

I was able to get a 100% accuracy on validation data, however it's likely due to a small dataset I did a 80/20 split for the dataset

Training