All Projects → laurahanu → Improved-Wasserstein-GAN-application-on-MRI-images

laurahanu / Improved-Wasserstein-GAN-application-on-MRI-images

Licence: other
Improved Wasserstein GAN (WGAN-GP) application on medical (MRI) images

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Improved-Wasserstein-GAN-application-on-MRI-images

2D-and-3D-Deep-Autoencoder
Convolutional AutoEncoder application on MRI images
Stars: ✭ 57 (+147.83%)
Mutual labels:  lasagne, theano, medical-imaging, mri-images, unsupervised-learning, medical-images
progressive growing of GANs
Pure tensorflow implementation of progressive growing of GANs
Stars: ✭ 31 (+34.78%)
Mutual labels:  generative-adversarial-network, wgan, wasserstein-gan, wgan-gp
Dcgan wgan wgan Gp lsgan sngan rsgan began acgan pggan tensorflow
Implementation of some different variants of GANs by tensorflow, Train the GAN in Google Cloud Colab, DCGAN, WGAN, WGAN-GP, LSGAN, SNGAN, RSGAN, RaSGAN, BEGAN, ACGAN, PGGAN, pix2pix, BigGAN
Stars: ✭ 166 (+621.74%)
Mutual labels:  generative-adversarial-network, wgan, wgan-gp
Tensorflow Generative Model Collections
Collection of generative models in Tensorflow
Stars: ✭ 3,785 (+16356.52%)
Mutual labels:  wgan, wgan-gp, improved-wgan
progressive-growing-of-gans.pytorch
Unofficial PyTorch implementation of "Progressive Growing of GANs for Improved Quality, Stability, and Variation".
Stars: ✭ 51 (+121.74%)
Mutual labels:  generative-adversarial-network, wgan, wasserstein-gan
Ganspapercollection
Stars: ✭ 130 (+465.22%)
Mutual labels:  theano, medical-imaging, generative-adversarial-network
Awesome Gans
Awesome Generative Adversarial Networks with tensorflow
Stars: ✭ 585 (+2443.48%)
Mutual labels:  generative-adversarial-network, wgan, wgan-gp
Pytorch Generative Model Collections
Collection of generative models in Pytorch version.
Stars: ✭ 2,296 (+9882.61%)
Mutual labels:  generative-adversarial-network, wgan, wgan-gp
WGAN-GP-tensorflow
Tensorflow Implementation of Paper "Improved Training of Wasserstein GANs"
Stars: ✭ 23 (+0%)
Mutual labels:  generative-adversarial-network, wgan, wgan-gp
All About The Gan
All About the GANs(Generative Adversarial Networks) - Summarized lists for GAN
Stars: ✭ 630 (+2639.13%)
Mutual labels:  medical-imaging, generative-adversarial-network, unsupervised-learning
Ganotebooks
wgan, wgan2(improved, gp), infogan, and dcgan implementation in lasagne, keras, pytorch
Stars: ✭ 1,446 (+6186.96%)
Mutual labels:  lasagne, wgan, wgan-gp
Alphazero gomoku
An implementation of the AlphaZero algorithm for Gomoku (also called Gobang or Five in a Row)
Stars: ✭ 2,570 (+11073.91%)
Mutual labels:  lasagne, theano
SemiDenseNet
Repository containing the code of one of the networks that we employed in the iSEG Grand MICCAI Challenge 2017, infant brain segmentation.
Stars: ✭ 55 (+139.13%)
Mutual labels:  theano, medical-imaging
Fun-with-MNIST
Playing with MNIST. Machine Learning. Generative Models.
Stars: ✭ 23 (+0%)
Mutual labels:  wgan, wgan-gp
mrnet
Building an ACL tear detector to spot knee injuries from MRIs with PyTorch (MRNet)
Stars: ✭ 98 (+326.09%)
Mutual labels:  medical-imaging, mri-images
Psgan
Periodic Spatial Generative Adversarial Networks
Stars: ✭ 108 (+369.57%)
Mutual labels:  lasagne, theano
Pytorch-Basic-GANs
Simple Pytorch implementations of most used Generative Adversarial Network (GAN) varieties.
Stars: ✭ 101 (+339.13%)
Mutual labels:  wgan, wgan-gp
rnn benchmarks
RNN benchmarks of pytorch, tensorflow and theano
Stars: ✭ 85 (+269.57%)
Mutual labels:  lasagne, theano
Repo 2016
R, Python and Mathematica Codes in Machine Learning, Deep Learning, Artificial Intelligence, NLP and Geolocation
Stars: ✭ 103 (+347.83%)
Mutual labels:  lasagne, theano
speech-enhancement-WGAN
speech enhancement GAN on waveform/log-power-spectrum data using Improved WGAN
Stars: ✭ 35 (+52.17%)
Mutual labels:  generative-adversarial-network, wgan-gp

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

alt text

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

alt text

  • Examples of generated images 7800 iterations

alt text

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

alt text

License

This project is licensed under Imperial College London.

Acknowledgements

The following codes were used as a base:

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].