Comparison of different CT metal artifact reduction strategies for standard titanium and carbon‐fiber reinforced polymer implants in sheep cadavers

Table 3 Distribution of streak artifacts and noise among different reconstruction algorithms and implant configurations (screw/rod). Distribution of most pronounced streak artifacts of the screw shaft (in ΔHU) and corresponding reference HU values of bone and fat tissue among different image reconstruction algorithms. Noise was measured as ± standard deviation (SD) of reference muscle attenuation. Note significantly lower artifact degrees for carbon (C) screw compared to titanium (Ti) screw containing configurations. Rod material had comparably lower impact on artifact degree. Mean HU-values of reference muscle tissue showed no significant differences among reconstructions-algorithms (F = 0.815, p = .615) while in the vertebral body significant inverse correlation with tube voltage (Spearman’s rho= −656, p < .001) was seen. Image noise inversely correlated with tube voltage

	Artifact degree (ΔHU) of screw shaft				Reference tissue attenuation (HU)
	Ti/Ti	Ti/C	C/C	C/Ti	Bone	Muscle	Noise (± SD)
sk SE 80 kV	− 859	− 903	− 37	− 32	872	63	19
bk SE 80 kV	− 870	− 898	− 35	− 58	871	63	45
sk SE 120 kV	− 848	− 897	− 11	− 39	670	60	14
bk SE 120 kV	− 876	− 920	− 48	− 69	656	61	37
sk SE 120 kV iMAR	− 768	− 719	− 71	− 106	628	59	16
sk DE 120 kV Mix	− 664	− 630	1	− 58	559	61	15
bk DE 120 kV Mix	− 713	− 674	− 1	− 63	562	61	43
sk SE Sn 150 kV	− 648	− 617	0	− 7	474	60	17
bk SE Sn 150 kV	− 691	− 641	− 10	− 5	477	60	53
sk DE ME 130 keV	− 615	− 582	13	− 17	402	59	18
bk DE ME 130 keV	− 690	− 608	11	− 25	401	60	46

bk bone kernel, C carbon, DE dual energy, HU Hounsfield Units, iMAR iterative metal artifact reduction (brand name), ME monoenergetic extrapolation, SD standard deviation, SE single energy, sk soft tissue kernel, Sn tin-filtered, Ti titanium

ISSN: 1471-2342