Sarabjeet Singh MBBS, Presenter: Nothing to Disclose

Mannudeep Karanvir Singh Kalra MD, Abstract Co-Author: Research grant, General Electric Company

Paul Licato MS, Abstract Co-Author: Employee, General Electric Company

Michael Austin Blake MBBCh, Abstract Co-Author: Nothing to Disclose

Jean-Baptiste Thibault, Abstract Co-Author: Employee, General Electric Company, Waukesha, WI

To assess objective noise and CT attenuation number attributes with standard and low dose abdominal CT examination reconstructed with model based iterative reconstruction (MBIR) and adaptive statistical iterative reconstruction (ASIR) technique.

5 patients (mean age 56 ± 6.8 years, M:F 4:1) gave informed consent for acquisition of 4 additional series through abdomen on a 64-slice MDCT (GE Discovery CT750 HD) at 200 (standard- SD), 150 (25%low dose- 25LD), 100 (50% low dose-50LD) and 50 (75% low dose- 75LD) mAs with remaining parameters held constant at 5mm slice thickness, 120 kVp and 0.984:1 pitch. Scan raw data were exported and images were reconstructed to generate ASIR and MBIR (GE Healthcare) 0.625 mm thick image series for all dose levels. Objective noise and HU were recorded in all ASIR and MBIR images in four different anatomic sites. Four point scale (1=none, 2 minor artifacts not interfering with diagnostic decision making, 3 major artifacts affecting visualization of major structures but diagnosis still possible, 4 affecting diagnostic information used to subjectively assess the presence of artifacts. Data were analyzed using analysis of variance (ANOVA).

Average CTDI vol for SD, 25LD, 50LD, and 75LD were 16.8 mGy, 12.6 mGy, 8.4 mGy and 4.2 mGy respectively (p<0.0001). For MBIR, there was no significant change in noise from SD to other three dose levels (p>0.05). Regardless of the dose levels there was no difference in CT numbers between ASIR and MBIR images in all four anatomic location (p=0.53-0.65). MBIR images had lower noise (25LD= 12.9% (16.3/18.7) lower noise, 50LD= 28.5% (16.9/23.7) lower noise, 75LD= 34.2% (19.4/29.4) lower noise) compared to ASIR images (p<0.0001). No significant artifacts were seen.

MBIR assisted image reconstruction of low dose abdominal CT examinations is associated with lower image noise without any significant image artifacts affecting diagnostic interpretation.

Ultra low dose CT scanning may be possible with MBIR assisted image reconstruction.

Singh, S, Kalra, M, Licato, P, Blake, M, Thibault, J, Assessment of Ultra-Low Dose Abdominal CT Reconstructed with Model-based Iterative Reconstruction and Adaptive Statistical Iterative Reconstruction: A Pilot Study.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8008874.html