Improved Wasserstein GAN (WGAN-GP) application on MRI dataset

Application of a deep generative adversarial network on MRI images of knees. The MRI database used was provided by Imperial College London, however similar databases can be found on the OAI website (http://www.oai.ucsf.edu/), an observational study dedicated to monitor the natural evolution of osteoarthritis.
The dataset used in this project was comprised of 28800 2D black&white MRI images of size 64x64.

Prerequisites

• Python, Lasagne (developer version), Theano (developer version), Numpy, Matplotlib, scikit-image
• NVIDIA GPU (5.0 or above)

Architecture

Results

The hyperparameters used were:

• a learning rate of 0.0005
• a decay rate of 0.5
• a batch size of 128 images
• a z space of 200

The network was trained using 35 epochs (~7800 iterations)

• Examples of real images from the input dataset

• Examples of generated images 7800 iterations

• Evolution of generated images at various iterations (total of 35 epochs - around 7800 iterations)

License

This project is licensed under Imperial College London.

Acknowledgements

The following codes were used as a base:

• for the main skeleton for a lasagne implementation of an adversarial generative network: https://github.com/ToniCreswell/AllThingsGAN/blob/master/Code/dcgan.py
• for WGAN-GP implementations
  https://github.com/tjwei/GANotebooks and
  https://github.com/igul222/improved_wgan_training