Bria Moore, Presenter: Nothing to Disclose

Daniel Tobias Boll MD, Abstract Co-Author: Research Grant, Siemens AG Research Grant, Koninklijke Philips NV Research Grant, Bracco Group

Juan Carlos Ramirez Giraldo PhD, Abstract Co-Author: Employee, Siemens AG

Joshua Wilson PhD, Abstract Co-Author: Nothing to Disclose

Ehsan Samei PhD, Abstract Co-Author: Research Grant, Siemens AG Research Grant, General Electric Company Research Grant, Carestream Health, Inc

Donald P. Frush MD, Abstract Co-Author: Nothing to Disclose

Artifact reduction via dual-energy CT (DECT) is predominantly studied in adults. This study aims to assess the impact of acquisition parameters on the reduction of a common pediatric hardware artifact, spinal rods, using a unique pediatric phantom.

A proprietary, custom-designed phantom (Mercury 3.0), consisting of a range of pediatric-sized modules (diameters of 12, 18.5, 23, and 30 cm) was fitted with two standard stainless steel rods used for pediatric scoliosis repair. The phantom was scanned with pediatric DECT protocols (Siemens Flash) using 80/140 kVp and 100/140 kVp respectively employing a dose modulated reference based on size. All studies were reconstructed using filtered back projection (FBP) and iterative reconstruction (IR, SAFIRE 3). DECT post processing (Syngo DE monoenergetic) was applied to reduce artifact in the reconstructed image by performing a monoenergetic extrapolation. Artifact reduction was assessed in the surrounding soft tissues by (1) comparing the attenuation between the corrected artifact series with that of the phantom in the absence of the artifact, and (2) evaluating the noise expressed as attenuation standard deviation between corrected image series with and without spinal rods.

Noise levels in the 100/140 kVp set were improved 20.7% on average when compared to the 80/140 kVp sets. IR reduced noise 26.3 % compared to FBP. For artifact reduction, the optimal keV settings (mean) for 100/140 kVp were similar (108 and 105 keV) for IR and FBP, respectively. However, the keV range was narrower for IR (101-109 keV) compared to that FBP (105-120 keV).  

The Mercury phantom is a unique tool for systematic investigation of pediatric CT, including innovative DECT applications. Our results suggest that optimal spinal rod metal artifact reduction in children using DECT is achieved from an acquisition protocol of 100/140 kVp pairs, iteratively reconstructed and DECT post-processed at 105-108 keV.

DECT can substantially reduce artifacts produced by common metal hardware such as spinal rods by using monoenergetic extrapolation post-processing.

Moore, B, Boll, D, Ramirez Giraldo, J, Wilson, J, Samei, E, Frush, D, Impact of Dual-Energy CT Techniques on Artifact Suppression Using a Pediatric Phantom.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045785.html