Bernhard Schmidt PhD, Presenter: Nothing to Disclose

Martin Ulrich Sedlmair MS, Abstract Co-Author: Support, Siemens AG, Forchheim, Germany

Bernhard Krauss PhD, Abstract Co-Author: Employee, Siemens AG, Forchheim, Germany

Jason Tyan, Abstract Co-Author: Employee, Siemens AG

Thomas G. Flohr PhD, Abstract Co-Author: Employee, Siemens AG, Forchheim, Germany

To investigate a quality assurance phantom for dual energy scanning and dual energy post-processing algorithms.

: With the recent introduction of two tube systems which allow dual energy scanning and with the various dual energy post processing methods on the market, the necessity for a quality assurance (QA) phantom is obvious to allow the comparison and the evaluation of the different available dual energy methods. The key applications for dual energy are the differentiation of materials (typically bone and iodine) and the calculation of pure iodine enhancement (I) and virtual non-contrast images (VNC) from contrast enhanced CT images. For our study we used an anthropomorphic chest phantom made out of tissue equivalent material (30 x 20 cmxcm) with a dedicated dual energy insert (diameter 10 cm). The insert contained lesions (1-4 cm) with different concentrations of iodine (200, 400, 600 HU @120 kV) and of calcium (200, 400, 600 HU @120 kV) simulating bone or calcified lesions. In addition two hypo dense lesions (CT value lowered by 140HU) were added, whereas one is enhanced by iodinated contrast agent to compensate at 120 kV the drop of CT values due to density. To simulate patients with a larger cross section, dedicated ‘fat’ rings were attached to the chest phantom to get up to patients cross sections of 40x30 cmxcm. Dual energy acquisitions were performed on a dual source CT system (SOMATOM Definition, Siemens Healthcare, Forchheim, Germany) with various acquisition setting and then dedicated post processing algorithms for the separation iodine/bone, quantification of iodine were applied (syngo Dual Energy SW, Siemens, Forchheim).

At 120kV there was no difference (< 5 HU) between regular tissue and hypo dense lesions with iodine. However, the DE post processed VNC images showed – as expected - a decrease by about 130 HU. At larger cross sections a dedicated beam hardening correction was necessary to compensate for an offset of about 10% at 400HU iodine enhancement at medium phantom size. For larger phantom diameters noise was challenging. The provided noise reduction filters helped to reduce the noise in the VNC images significantly (about factor of 2) without affecting absolute CT values. The separation of iodine and bone also worked for higher noise levels and also impacted image quality.

For regular patient diameters both, lesions with iodine and calcium were labeled correctly. Also the separation of hypo-dense lesions with iodine and regular tissue was possible. However, especially for larger patient diameter’s, the adaptation of tube current is recommended e.g. AEC and dedicated image filers to compensate for beam hardening and noise in the post processed images are necessary.

The investigated dual energy QA phantom helps to assess the quality of DE images and the accuracy of DE post-processing algorithms.

Schmidt, B, Sedlmair, M, Krauss, B, Tyan, J, Flohr, T, Assessment of a Quality Assurance Phantom for Dual Energy CT.  Radiological Society of North America 2008 Scientific Assembly and Annual Meeting, February 18 - February 20, 2008 ,Chicago IL. http://archive.rsna.org/2008/6018424.html