Ripple Artifact Quantification in Slice Encoding for Metal Artifact Correction (SEMAC) using MR Bloch Simulation

Motivation: Commonly used methods for the quantification of metal-induced image artifacts, such as measuring the extent of signal voids, do not capture spatial frequencies of ripple artifacts, as seen in dedicated metal artifact reduction sequences including SEMAC or MAVRIC. Goal(s): To propose a new method for the quantification of SEMAC ripple artifacts which may serve as quality metric for sequence optimizations. Approach: We applied a k-space-based metric to MR Bloch simulations of SEMAC sequences with variable slice thicknesses and RF pulse shapes (time-bandwidth product, TBW). Results: A trend towards higher absolute artifact intensity and lower spatial frequency can be observed for higher TBWs. Impact: The proposed metal artifact metric extends current quantification methods by taking the spatial frequency distribution of ripple artifacts into account. This may serve as a basis for metal artifact reduction sequence optimization, with a particular focus on RF pulse parameters.