Algorithm 1. The VGG-19 model description.

Input: CXR images of dimension 500 × 500 pixels from the Training CXR Dataset Output: VGG model weights 1. epochs ← 100 2. for each image in the dataset do 3.  resize the image to 224 × 224 pixels 4.  normalize the image pixels values from (0,255) to (0,1) 5. end 6. Load the VGG-19 model pre-trained on the ImageNet dataset 7. Remove the last layer of the model 8. Make non-trainable all the layers of the model 9. Add a Flatten layer to the model output to obtain 1-D array of features 10. Apply a batch normalization to the 1-D array of features 11. Add a fully connected layer with 256 hidden neurons 12. Apply a dropout for inactivate units (40%) in the previous layer 13. Add a fully connected layer with 128 hidden neurons 14. Apply a dropout for inactivate units (60%) in the previous layer 15. Apply a batch normalization 16. Add a fully connected layer with four hidden units and a softmax activation function. 17. Optimize the model with Adam with learning_rate = 0.01 and a decay = learning_rate/epochs 18. Train the model for the given number of epochs and a batch size of 32 19. Save the final model