eduzhai > Applied Sciences > Engineering >

Deep Parallel MRI Reconstruction Network Without Coil Sensitivities

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 13 pages

Abstract: We propose a novel deep neural network architecture by mapping the robustproximal gradient scheme for fast image reconstruction in parallel MRI (pMRI)with regularization function trained from data. The proposed network learns toadaptively combine the multi-coil images from incomplete pMRI data into asingle image with homogeneous contrast, which is then passed to a nonlinearencoder to efficiently extract sparse features of the image. Unlike most ofexisting deep image reconstruction networks, our network does not requireknowledge of sensitivity maps, which can be difficult to estimate accurately,and have been a major bottleneck of image reconstruction in real-world pMRIapplications. The experimental results demonstrate the promising performance ofour method on a variety of pMRI imaging data sets.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×