Meeting Banner
Abstract #2584

Robust Low-Rank Matrix Completion for Sparse Motion Correction in Auto Calibration PI

Zhongyuan Bi1, Martin Uecker2, Dengrong Jiang3, Michael Lustig2, Kui Ying3

1Biomedical Engineering, Tsinghua University, Beijing, China; 2Electrical Engineering and Computer Science, University of California Berkeley, Berkeley, CA, United States; 3Engineering Physics, Tsinghua University, Beijing, China

Auto-calibration parallel imaging (acPI) is based on local correlations in k-space. It is known to perform robustly in practice, especially when accurate sensitivity information is hard to obtain. However, corruption of ACS data, e.g. by motion, often leads to serious artifacts in the reconstructed images. In this work, we propose to exploit the redundancy in k-space to detect and correct sparse corruptions in ACS data, which could result from random, time-limited motion in clinical practice (e.g. swallowing, jerk, etc). Our work is based on low-rank matrix completion with sparse errors.

Keywords

accuracy accurate achieved added almost appear apply approximation artifacts assumption auto axial biomedical blocks boundaries calibration caused channel china clinical coil completed completion computational compute computer consistency construct control convergence correct corrected correction correlation correlations corrupted corruption corruptions costs dependent described detect distributed done efficiency efficiently electrical enforce engineering entries errors especially every exists exploit expressed extension flowchart fully general gradient greatly hard head highly identical improve in vivo incoherent initial interpolation intuitively inversion jerk kernels known lapping leading leads like limited local martin matrix missing motion motions newly often onto opportunities original parallel physics positions potential practice prepared preserved problems propose proposed quality random randomly rank reconstruct reconstructed reconstruction recovery reduce reduced redundancy regularization renders repeat residual respectively rest robust robustly scheme science sensitivity serious setup shrink sign simulate simulation singular situations soft solve space sparse spoiled stand still swallow swallowing synthesized system threshold thresholding till uncorrupted underway verifies verify whole wise