Skip to content

Implementation code for "Iterative PET Image Reconstruction Using Convolutional Neural Network Representation"

License

Notifications You must be signed in to change notification settings

suezhao/IterativeCNN

 
 

Repository files navigation

Iterative PET Image Reconstruction Using CNN Representation

result

Prerequisites

Matlab, Python 2.7, Tensorflow 1.4

Instructions

Note

  • The 3D U-net is trained and the trained model is stored in pretraining_process. If you want to re-train the model based on your own data sets, you can use Unet3D_train.py for training and Unet3D_test.py for testing.
  • The results runing on Ubuntu server setting penalty parameter rho = 7.5e-4 are uploaded for reference.
  • If you do not have training data, but have the anatomical prior image, you can check my newly published DIPRecon method.

Reference

Gong, K., Guan, J., Kim, K., Zhang, X., Fakhri, G.E., Qi, J.* and Li, Q.*, 2017. Iterative PET image reconstruction using convolutional neural network representation. arXiv preprint arXiv:1710.03344.
Gong, K., Guan, J., Kim, K., Zhang, X., Yang, J., Seo, Y., Fakhri, G.E., Qi, J.* and Li, Q.*, 2018. Iterative PET image reconstruction using convolutional neural network representation. IEEE Transactions on Medical Imaging

License

This project is licensed under the 3-Clause BSD License - see the LICENSE file for details.

About

Implementation code for "Iterative PET Image Reconstruction Using Convolutional Neural Network Representation"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 69.4%
  • MATLAB 30.6%