%0 Conference Paper
%A Dedmari, Muneer Ahmad
%A Conjeti, Sailesh
%A Estrada Leon, Edgar Santiago
%A Ehses, Philipp
%A Stöcker, Tony
%A Reuter, Martin
%T Complex Fully Convolutional Neural Networks for MR Image Reconstruction
%M DZNE-2020-00874
%@ 978-3-030-00128-5 (print)
%D 2018
%X Undersampling the k-space data is widely adopted for acceleration of Magnetic Resonance Imaging (MRI). Current deep learning based approaches for supervised learning of MRI image reconstruction employ real-valued operations and representations by treating complex valued k-space/spatial-space as real values. In this paper, we propose complex dense fully convolutional neural network (CDFNet) for learning to de-alias the reconstruction artifacts within undersampled MRI images. We fashioned a densely-connected fully convolutional block tailored for complex-valued inputs by introducing dedicated layers such as complex convolution, batch normalization, non-linearities etc. CDFNet leverages the inherently complex-valued nature of input k-space and learns richer representations. We demonstrate improved perceptual quality and recovery of anatomical structures through CDFNet in contrast to its real-valued counterparts.
%B Machine Learning for Medical Image Reconstruction
%C 16 Sep 2018 - 16 Sep 2018, Granada (Spain)
Y2 16 Sep 2018 - 16 Sep 2018
M2 Granada, Spain
%F PUB:(DE-HGF)3 ; PUB:(DE-HGF)6
%9 BookConference Presentation
%R 10.1007/978-3-030-00129-2_4
%U https://pub.dzne.de/record/145540