Experimental Study of MAR Algorithm in Metal Artifact Removal of CT Simulator

Objective To explore the application value of MAR algorithm in metal artifact removal of CT simulator. Methods CT phantom with titanium plate was scanned using conventional algorithms and MAR algorithms, respectively. Artifact index(AI), contrast-to-noise ratio(CNR) and AI values at different slices were used to analyze the artifact images. Results In artifact index, MAR algorithm (10.28±2.60) is significantly lower than conventional algorithm (20.65±5.04); In contrast-to-noise ratio index, MAR algorithm (7.81±1.12) is better than conventional algorithm (5.61±1.36). The above indicators were statistically significant in both algorithms (P<0.01). In the slices affected by metal artifacts, the artifact index decreased by 21.72%~88.40% after the MAR algorithm. Conclusion MAR algorithm can significantly reduce the metal artifacts and improve the clinical value of CT data.