Haruhiko Machida MD, Presenter: Nothing to Disclose

Hitoshi Takeuchi MD, Abstract Co-Author: Nothing to Disclose

Isao Tanaka, Abstract Co-Author: Nothing to Disclose

Rika Fukui, Abstract Co-Author: Nothing to Disclose

Xiao Zhu Lin BA, Abstract Co-Author: Nothing to Disclose

Eiko Ueno MD, Abstract Co-Author: Nothing to Disclose

Shigeru Suzuki, Abstract Co-Author: Nothing to Disclose

Kazufumi Suzuki MD, Abstract Co-Author: Nothing to Disclose

Masami Hirata, Abstract Co-Author: Nothing to Disclose

Shinya Kojima, Abstract Co-Author: Nothing to Disclose

Yun Shen PhD, Abstract Co-Author: Employee, General Electric Company Researcher, General Electric Company

To investigate whether arteries of the brain are better depicted by CT angiography (CTA) using model-based iterative reconstruction (MBIR) than CTA using filtered back projection (FBP), even in the posterior cranial fossa (PCF).

We reconstructed axial images of 28 consecutive patients (14 men, 14 women; mean age 58.6 ± 14.6 years) who underwent CTA of the brain (tube voltage, 120 kV; tube current, 335 mA; rotation speed, 0.5 sec; pitch, 0.516) with a 64-detector CT scanner (Discovery CT750 HD, GE). With volume-rendering CTA, we used FBP with standard kernel and MBIR to reconstruct images 0.625-mm thick. To measure the contrast-to-noise ratio (CNR), we placed regions of interest on the axial images within the pons and bilateral cerebellar hemisphere and assessed the mean CT value and standard deviation (SD) (noise) in the PCF and basilar artery (BA) using the equation: (mean CT value in BA – mean CT value in PCF) / noise in PCF. Using the CTA images and a 4-point scale (1: poor; 4 excellent), 2 radiologists independently graded delineation of the arteries in PCF, consisting of the BA, bilateral vertebral artery (VA), and their branches−the bilateral superior cerebellar artery (SCA) and anterior (AICA) and posterior inferior cerebellar artery (PICA). We compared results between FBP and MBIR using Wilcoxon signed-rank test.

The CNR was significantly higher by MBIR (48.2 ± 10.1) than FBP (37.9 ± 11.0) (P < 0.0001). Subjective arterial delineation was similar for MBIR and FBP in the BA (Reader 1: 4.0 ± 0.2 vs 3.9 ± 0.4; Reader 2: 4.0 ± 0.0 vs 3.9 ± 0.3; P > 0.05 for both) but significantly higher for MBIR in the VA, SCA, AICA, and PICA (Reader 1: 4.0 ± 0.1 vs 3.9 ± 0.3 [VA]; 3.0 ± 0.6 vs 2.1 ± 0.7 [SCA]; 2.6 ± 0.7 vs 1.6 ± 0.7 [AICA]; 3.3 ± 0.5 vs 2.7 ± 0.6 [PICA]; Reader 2: 4.0 ± 0.0 vs 3.9 ± 0.3 [VA]; 3.6 ± 0.7 vs 2.5 ± 0.9 [SCA]; 3.2 ± 0.9 vs 1.7 ± 0.9 [AICA]; 3.8 ± 0.4 vs 2.9 ± 0.8 [PICA]; P < 0.05 for all).

With CTA of the brain, CNR and arterial delineation are better using MBIR than FBP, even in the PFC, as shown in the VA, SCA, AICA, and PICA.

With CTA of the brain, MBIR may provide better image quality and arterial delineation than FBP, especially for small branches even in the posterior cranial fossa, susceptible to artifact.

Machida, H, Takeuchi, H, Tanaka, I, Fukui, R, Lin, X, Ueno, E, Suzuki, S, Suzuki, K, Hirata, M, Kojima, S, Shen, Y, Brain Arteries Are Better Delineated in CT Angiography Using Model-based Iterative Reconstruction than Filtered Back Projection.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11020104.html