hanyoseob/framing-u-net

Deep Convolutional Framelets: A General Deep Learning Framework for Inverse Problems

Paper

• Deep Convolutional Framelets: A General Deep Learning Framework for Inverse Problems
  • In press in SIAM Journal on Imaging Sciences (2018): [https://arxiv.org/abs/1707.00372]
• Framing U-Net via Deep Convolutional Framelets: Application to Sparse-view CT
  • In revision process: [https://arxiv.org/abs/1708.08333]

Implementation

• MatConvNet (matconvnet-1.0-beta24)
  • Please run the matconvnet-1.0-beta24/matlab/vl_compilenn.m file to compile matconvnet.
  • There is instruction on "http://www.vlfeat.org/matconvnet/mfiles/vl_compilenn/"
• Frameing U-Net (matconvnet-1.0-beta24/examples/framing_u-net)
  • Please run the matconvnet-1.0-beta24/examples/framing_u-net/install.m
  • Install the customized library
  • Download the trained networks such as standard cnn, u-net, and tight-frame u-net

Trained network

• Trained network for 'Standard CNN' is uploaded.
• Trained network for 'U-Net' is uploaded.
• Trained network for 'Tight-frame U-Net' is uploaded.

Test data

• Iillustate the Fig. 5 for Framing U-Net via Deep Convolutional Framelets:Application to Sparse-view CT
• CT images from '2016 Low-Dose CT Grand Challenge' are uploaded to test.
  • Thanks Dr. Cynthia McCollough, the Mayo Clinic, the American Association of Physicists in Medicine(AAPM), and grand EB017095 and EB017185 from the National Institute of Biomedical Imaging and Bioengineering for providing the Low-Dose CT Grand Challenge dataset.