| Paper ID | B-1-3.1 |
| Paper Title |
Deep-Learning-based MR Compressed Sensing using Non-randomly Under-sampled Signal in Nonlinear Phase Encoding Imaging |
| Authors |
Satoshi ITO, Shohei OUCHI, Utsunomiya University, Japan |
| Session |
B-1-3: Signal Processing in Medical/Clinical Sciences |
| Time | Tuesday, 08 December, 17:15 - 19:15 |
| Presentation Time: | Tuesday, 08 December, 17:15 - 17:30 Check your Time Zone |
|
All times are in New Zealand Time (UTC +13) |
| Topic |
Biomedical Signal Processing and Systems (BioSiPS): |
| Abstract |
Optional image scaling, and hence aliasless image reconstruction, is feasible using a signal that violates the sampling theorem in MR phase scrambling Fourier transform imaging. In this method, the main and aliased image components are separated in the scaled space when a large scaling factor is selected. In the present study, a new fast imaging method, in which aliasing artifacts caused by undersampling of the signal, are removed in two steps: in the downscaled space introduced by aliasless reconstruction and through de-aliasing using a deep convolution neural network. The proposed method is shown to provide higher PSNR images compared to random sampling compressed sensing and has an advantage in terms of low-sampling-rate image acquisition. |