Artifacts-Free Multi-Shot Liver Diffusion Weighted Imaging with Phase Constrained Reconstruction

Motivation: State-of-the-art navigation-free methods encounter problems such as difficulty in removing motion artifacts and excessive noise in liver diffusion-weighted imaging (DWI) reconstruction. Goal(s): To propose a rapid and high-resolution reconstruction method that can eliminate motion artifact effectively for liver multi-shot DWI. Approach: We provide a one-dimensional low-rank reconstruction algorithm based on k-space separation guided reconstruction for high-resolution liver multi-shot DWI. Results: ONe-dimensional mIxed-domain low-rank Constrained recoverY (NICY) extends one-dimensional low-rank constraint to a two-dimensional image reconstruction framework, not only significantly reduces the adverse effects of liver motion on imaging quality, effectively eliminating the resulting artifacts, but also enhances the image's SNR. Impact: The propose method can eliminate motion artifacts and improve the SNR of reconstructed images. The successful application of this method is expected to enable liver DWI to play a greater role in early diagnosis and efficacy evaluation of clinical diseases.