Comparison of the diagnostic accuracy of FBP, ASiR, and MBIR reconstruction during CT angiography in the evaluation of a vessel phantom with calcified stenosis in a distal superficial femoral artery in a cadaver extremity.

Purpose: To investigate whether adaptive statistical iterative reconstruction (ASiR) or model-based iterative reconstruction (MBIR) improves the diagnostic performance of computed tomography angiography (CTA) for small-vessel calcified lesions relative to filtered back projection (FBP) using cadaver extremities and a calcified stenosis phantom. Methods: A cadaver was used in accordance with our institutional regulations, and a calcified stenosis phantom simulating 4 grades of stenosis was prepared. The phantom was inserted within the distal superficial femoral artery of the cadaver leg. Ten CT images per reconstruction type and stenosis grade were acquired using a 64-slice multidetector-row CTA. As an objective measurement, the first and second derivatives of the CT value function profiles were calculated. As a subjective measurement, 2 blinded reviewers measured the stenosis ratio using a quantitative scale. The Wilcoxon rank-sum test was used to evaluate the data. Results: Objective measurements of both 25% and 50% stenosis differed significantly (P<0.01) between MBIR (25/50%: 25.80/50.30 +/- 3.88/3.86%) and FBP (25/50%: 35.60/83.80 +/- 3.44/26.10%), whereas significant differences were not observed between ASiR and FBP. Reviewer 2's subjective measurements of 25% stenosis differed significantly (P<0.01) between MBIR (35.13 +/- 3.25%) and ASiR (40.89 +/- 3.14%), and the measurements of 50% stenosis differed significantly (P<0.01) between MBIR (reviewers 1/2, 62.36/54.78 +/- 2.78/4.96%) and FBP (reviewers 1/2, 62.36/74.84 +/- 2.78/18.10%). Significant differences in the subjective measurements were not observed between ASiR and FBP. Conclusion: MBIR improves the diagnostic performance of CTA for small-vessel calcified lesions relative to FBP.