Combined use of virtual monochromatic images and projection-based metal artifact reduction methods in evaluation of total knee arthroplasty.

Abstract

To evaluate the effect of the combined use of virtual monochromatic imaging (VMI) and projection-based metal artifact reduction (PB-MAR) methods on further artifact reduction and image quality in total knee arthroplasty (TKA) patients. Fifty-seven knee joints from 36 patients who had a previous history of TKA for bilateral or unilateral knee joints were included in this study. Four sets of images were compared between non-MAR, PB-MAR, VMI, and VMI + PB-MAR. For quantitative analysis, the area, mean attenuation, artifact index (AI), and contrast-to-noise ratio (CNR) were calculated for each protocol. Regarding qualitative analysis, overall artifact and depiction of soft tissue and bony structure were compared using relative visual grading analysis. In the femoral region, the VMI + PB-MAR protocol showed the best MAR performance in quantitative measures including area, mean attenuation, and AI (p < .001). However, MAR protocols with VMI showed significantly lower CNRs than did the protocols without VMI in the tibial region (p < .001). On qualitative analysis, VMI + PB-MAR also showed fewer overall artifacts than did the other two MAR protocols. Soft tissue was best depicted in VMI + PB-MAR protocol, and PB-MAR showed the best performance for the depiction of bony structure (p < .001). The combined use of VMI and PB-MAR showed better MAR performance than did PB-MAR or VMI alone. In areas with severe metal artifacts, the VMI + PB-MAR and PB-MAR protocols were useful for the evaluation of soft tissue and bone structure, respectively. • The combination of artifact reduction methods is effective in reducing metal artifacts in CT. • Soft tissue was best depicted in the combined protocol of projection- and dual-energy-based methods, and projection-based protocol showed the best performance for the depiction of bone structure in case of severe metal artifacts.