Abstract #3598

Attention Based Scale Recurrent Network for Under-Sampled MRI Reconstruction

Gabriel della Maggiora^1,2,3, Alberto Di Biase^1,2,3, Carlos Castillo-Passi^2,3,4, and Pablo Irarrazaval^1,2,3,4

¹Electrical Engineering Department, Pontificia Universidad Católica de Chile, Santiago, Chile, ²Biomedical Imaging Center, Pontificia Universidad Católica de Chile, Santiago, Chile, ³Millennium Nucleus for Cardiovascular Magnetic Resonance, Santiago, Chile, ⁴Institute for Biological and Medical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile

We propose an Attention Based Scale Recurrent Network for reconstructing under-sampled MRI data. This network is a variation of the recently proposed Scale Recurrent Network for blind deblurring¹. We treat the reconstruction problem as a deblurring problem. Thus the under-sampling pattern does not need to be known. We trained and tested our network with the NYU knee dataset available for the fastMRI challenge. The proposed model shows promising results for single-coil reconstruction outperforming both baselines.

This abstract and the presentation materials are available to members only; a login is required.

Join Here