Compressive Slice Encoding for Metal Artifact Correction

Introduction: Slice encoding for metal artifact correction (SEMAC) [1] completely corrects metal-induced in-plane and through-plane distortions. The SEMAC imaging sequence (Fig. 1) augments a view-angle-tilting (VAT) [2] spin echo sequence with z-phase encoding. While the VATcompensation gradient suppresses the in-plane distortions, the throughplane distortions are corrected by combining multiple 3D slabs, whose distorted excitation profiles are resolved with the z-phase encoding. However, the additional z-phase encoding leads to longer scan times. To accelerate SEMAC acquisition, the SEMAC imaging sequence has been integrated with parallel imaging and partial Fourier acquisition along the y phase encoding direction [3], which nonetheless involves a tradeoff of signal-to-noise ratio (SNR). In this work SEMAC is incorporated with compressed sensing (CS) [4] to reduce both y and z phase encoding. We demonstrate that the new technique, referred to as Compressive SEMAC, can greatly reduce scan times, while producing high-quality distortion correction and SNR comparable to SEMAC with full sampling. Methods and Results: Compressive SEMAC randomly skips both y and z phase encoding steps with a variable-density sampling pattern in ky − kz space. The excitation profile in a 3D slab is recovered by performing the following constrained reconstruction at each x location: min ||myz ||1 +λTV(myz ) subject to || Fu myz −Y ||< ε (1)