Laplacian Pyramid-based Complex Neural Network Learning for Fast MR - PowerPoint PPT Presentation

Laplacian Pyramid-based Complex Neural Network Learning for Fast MR Imaging Haoyun Liang Paul C. Lauterbur Research Center for Biomedical Imaging, SIAT, CAS

Accelerating MRI Fast Imaging Sequence Parallel Imaging Compressed Sensing Deep Learning

Motivations • multi-scale properties are underutilized • the blurring issue of textures and details of tissues and organs • normal convolution can not make full use of the information in complex-valued MR images

Contributions • pyramid structure decomposition is introduced to leverage multi- scale properties • cascaded structure is used for better restore textures and details of the reconstructed images • complex convolution is introduced to make full use of the information in complex-valued MR images

Network Structure 4x Shuffle downsample 4x Shuffle upsample 2x Linear upsample 2x Shuffle upsample Conv Block - 1 Conv Block - 2 Complex Laplacian pyramid decomposition Per-pixel Conv Add

Network Structure r X Shuffle downsample r X Shuffle upsample … ! #$% ! !" ! #$% Residual Block ! !" … ℎ #$% ℎ #$% … ℎ !" ℎ !" # #$% # !" Complex conv ! #$% = ! !" //r ! #$% = ! !" ∗ r Complex conv # #$% # !" ℎ #$% = ℎ !" //r ℎ #$% = ℎ !" ∗ r ReLU ReLU # #$% = # !" ∗ & & # #$% = # !" //& & Add

Result Reference Under-sampling U-Net KIKI-Net Cascade-Net CLP-Net kspace learning

Result Reference Under-sampling U-Net KIKI-Net Cascade-Net kspace learning CLP-Net