How Well Does Dual-Energy Computed Tomography With Metal Artifact Reduction Software Improve Image Quality and Quantify Computed Tomography Number and Iodine Concentration?

OBJECTIVE: The objective of this study was to assess the accuracy of the quantitative measurements obtained using dual-energy computed tomography with metal artifact reduction software (MARS). METHODS: Dual-energy computed tomography scans (fast kV-switching) are performed on a phantom, by varying the number of metal rods (Ti and Pb) and reference iodine materials. Objective and subjective image analyses are performed on retroreconstructed virtual monochromatic images (VMIs) (VMI at 70 keV). RESULTS: The maximum artifact indices for VMI-Ti and VMI-Pb (5 metal rods) with MARS (without MARS) were 17.4 (166.7) and 34.6 (810.6), respectively; MARS significantly improved the mean subjective 5-point score (P < 0.05). The maximum differences between the measured Hounsfield unit and theoretical values for 5 mg/mL iodine and 2-mm core rods were -42.2% and -68.5%, for VMI-Ti and VMI-Pb (5 metal rods), respectively, and the corresponding differences in the iodine concentration were -64.7% and -73.0%, respectively. CONCLUSIONS: Metal artifact reduction software improved the objective and subjective image quality; however, the quantitative values were underestimated.