This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Cardiac, Cardiovascular system, CT-Angiography, CT, Comparative studies, Computer Applications-Detection, diagnosis, Image verification
Authors:
H. Eom, D. H. Yang, Y.-H. Kim, J. Kwon, J.-H. Roh, J.-W. Kang, T.-H. Lim; Seoul/KR
DOI:
10.1594/ecr2015/C-1924
Results
Phantom
Crushed segment appeared as denser and thicker than non-crushed segment on dual source CT at the junction of MB and SB stent on either side (figure 2).
Floating strut appeared as protruding of the stent on dual source CT (figure 2).
The classification of the location of crushed segment,
presence of floating strut and unintended gap matched between micro-CT and dual source CT.
Among 9 stents,
8 (89%) were central type,
4 (44%) had floating strut and 2 (22%) showed unintended gap.
Parameters measured in both micro-CT and dual source CT showed good correlation.
According to the Bland-Altman plots of agreement,
the length and perimeter of crushed segment and overlapped segment on dual source CT showed high agreement with micro CT (95% limits of Agreement = -3.2 to 3.8 mm; 95% CI =-0.74 to 1.32,
95% limits of Agreement = -2.5 to 3.6 mm; 95% CI =-0.50 to 1.57) (figure 5).
Also,
the minimal luminal area at the level of crushed segment,
floating strut and overlapped segment,
SB ostium,
and non-crushed SB segment on dual source CT showed high agreement with micro CT (95% limits of Agreement = -1.5 to 1.6 mm; 95% CI = -0.22 to 0.26) (figure 5).