Yan Xing PhD, MD, Presenter: Nothing to Disclose

wen ya liu, Abstract Co-Author: Nothing to Disclose

Cunxue Pan PhD, Abstract Co-Author: Nothing to Disclose

Gulina Azhati, Abstract Co-Author: Nothing to Disclose

Jun Dang, Abstract Co-Author: Nothing to Disclose

jing jing LI, Abstract Co-Author: Nothing to Disclose

Haiting Ma, Abstract Co-Author: Nothing to Disclose

Yan Wei Wang MD, Abstract Co-Author: Nothing to Disclose

To investigate the feasibility of a novel intelligent boundary registration (IBR) technique to align stair-step artifacts in coronary CT angiography (CCTA).

Twenty-one consecutive CCTA exams with varying degrees of coronary artery stair-step artifacts were retrospectively processed with IBR technique on workstation (Advantage Windows 4.6; GE Healthcare). Two observers evaluate stair-step artifacts on IBR on and off images on per-segment basis defined by the 15-segment American Heart Association (AHA) guidelines. The severity of stair-step artifacts was graded with a 5-point grading scale (1.severe, complete discontinuity of the proximal and distal portions of the coronary artery; 2.moderate, discontinuity >50% of the artery diameter; 3.slight, discontinuity 25%–50% of the diameter; 4.minimal, discontinuity <25% of the diameter; and 5.no stair-step artifact). Images scored 1 or 2 were considered non-assessable. Comparisons of variables were performed with Wilcoxon rank sum test and McNemar test.

A total of 50 stair-step artifacts were found (35 Right Coronary Artery, 12 Left Coronary Artery, 3 Left Circumflex Artery). Images with IBR on processing were rated as significantly higher image scores versus those with IBR off (Average Image score: 4.42 ±1.13 vs 2.94 ±1.10);  (Z=5.681, P=0.000). Stair-step artifacts were fully corrected (Image score = 5 with IBR) in 70% (35/50) of all segments. Images with IBR off processing were rated as significantly higher non-assessable segments versus those with IBR on (Non-assessable rate: 34% vs 10% ); (χ2=8.392, P=0.004).

This novel IBR technique is feasible to reduce the severity of stair-step artifacts and increase assessable segments in CCTA.

The use of IBR technique may reduce the number and severity of stair-step artifacts in CCTA, potentially increasing diagnostic confidence.  

Xing, Y, liu, w, Pan, C, Azhati, G, Dang, J, LI, j, Ma, H, Wang, Y, Initial Experience of Intelligent Boundary Registration in Coronary CTA.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14010619.html