Annihilating filter-based low rank Hankel matrix approach (ALOHA) was recently used as a reference-free ghost artifact correction method. Inspired by another discovery that convolutional neural network can be represented by Hankel matrix decomposition, here we propose a deep CNN for reference-free EPI ghost correction. Using real EPI experiments, we demonstrate that the proposed method effectively removes the ghost artifacts with much faster reconstruction time compared to the existing reference-free approaches.
This abstract and the presentation materials are available to members only; a login is required.