Premium
Robust SENSE reconstruction of simultaneous multislice EPI with low‐rank enhanced coil sensitivity calibration and slice‐dependent 2D Nyquist ghost correction
Author(s) -
Lyu Mengye,
Barth Markus,
Xie Victor B.,
Liu Yilong,
Ma Xin,
Feng Yanqiu,
Wu Ed X.
Publication year - 2018
Publication title -
magnetic resonance in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.696
H-Index - 225
eISSN - 1522-2594
pISSN - 0740-3194
DOI - 10.1002/mrm.27120
Subject(s) - sensitivity (control systems) , calibration , sense (electronics) , multislice , rank (graph theory) , physics , nuclear magnetic resonance , nyquist–shannon sampling theorem , electromagnetic coil , nyquist frequency , mathematics , computer science , optics , artificial intelligence , algorithm , computer vision , chemistry , statistics , electronic engineering , combinatorics , engineering , quantum mechanics , filter (signal processing)
Purpose To improve simultaneous multislice (SMS) EPI by robust Nyquist ghost correction in both coil sensitivity calibration and SMS reconstruction. Methods To derive coil sensitivity and slice‐dependent phase difference map between positive‐ and negative‐echo images, single‐band EPI reference data are fully sampled with EPI parameters matched to SMS acquisition. First, the reference data are organized into positive‐ and negative‐echo virtual channels where missing data are estimated using low‐rank‐based simultaneous autocalibrating and k‐space estimation (SAKE) at small matrix size. The resulting ghost‐free positive‐ and negative‐echo images are combined to generate coil sensitivity maps. Second, full‐matrix positive‐ and negative‐echo images are SENSE reconstructed from the reference data. Their phase difference or error map is then calculated. Last, SMS EPI is reconstructed using phase error correction SENSE (PEC‐SENSE) that incorporates phase error map into coil sensitivity maps for negative‐echo data. The proposed method was evaluated using both experimental data from 7T systems and simulations. Results Virtual coil SAKE eliminated Nyquist ghosts in the single‐band EPI, yielding high‐quality coil sensitivity maps and phase error maps. The subsequent PEC‐SENSE robustly reconstructed SMS EPI under various conditions, including presence of in‐plane acceleration, with lesser artifacts and higher temporal SNR than slice‐dependent 1D linear correction method. Conclusion The proposed procedure of virtual coil SAKE calibration and PEC‐SENSE reconstruction substantially reduces all ghost‐related artifacts originating either directly from SMS EPI data or indirectly from EPI‐based coil sensitivity maps. It is computationally efficient, and generally applicable to all SMS EPI‐based applications.