Marc Kachelriess PhD, Presenter: Nothing to Disclose

Michael Knaup PhD, Abstract Co-Author: Nothing to Disclose

Christian Penssel PhD, Abstract Co-Author: Nothing to Disclose

Thomas G. Flohr PhD, Abstract Co-Author: Nothing to Disclose

Karl Stierstorfer PhD, Abstract Co-Author: Nothing to Disclose

Willi A. Kalender PhD, Abstract Co-Author: Nothing to Disclose

The sampling theorem demands two data points per resolution element. Our 64-slice CT scanner is equipped with a zFFS that allows to acquire two slices per detector row. We evaluated this zFFS regarding artifact behavior and image quality.

The Sensation 64 (Siemens Medical Solutions, Forchheim, Germany) 64-slice scanner is equipped with a 40-row detector. Its central 32 rows acquire 0.6 mm slices and a zFFS switches back and forth between two z-positions between each reading. Two slices per detector row are acquired and the scanner fulfills the Nyquist sampling theorem in the z-direction and promises to show less spiral windmill artifacts. To evaluate the zFFS (which cannot be switched off) we implemented a modified ASSR reconstruction that is able to skip every other view and thus obtains images as they would be obtained without zFFS. Averaging two of these reconstructions yields the correct zFFS-reconstructions. To match image noise (2320 vs. 1160 readings per rotation) random noise was added to the zFFS-reconstruction. A patient study consisting of 5 abdomen, 5 thorax and 5 head exams was used to analyze the influence of the focal spot deflection. Since z-sampling becomes more critical with larger pitch values we further performed scans of an artifact provoking phantom and performed a cadaver study with pitch values ranging from 0.5 to 1.5.

We found significantly increased windmill artifacts without the zFFS. Especially high contrast objects (bones in the head and pelvis and the ribs in the thorax region) suffer from these subtle streak artifacts that appear to rotate around the object when scrolling through the slices. Our phantom and cadaver studies that analyze the artifact behavior as a function of increasing pitch showed increasing artifact level for the modified reconstruction and constantly low artifacts for the standard Sensation 64 reconstruction.

The focal spot deflection helps to significantly reduce spiral artifacts. This shows that these artifacts are not a cone-beam problem as it is often believed. Obviously, sampling and aliasing play a significant role, and excellent image quality can only be achieved if this is taken into account explicitly.

T.G.F.,K.S.: Thomas Flohr and Karl Stierstorfer are with Siemens Medical Solutions, Forchheim, Germany.

Kachelriess, M, Knaup, M, Penssel, C, Flohr, T, Stierstorfer, K, Kalender, W, Evaluation of the z-Flying Focal Spot (zFFS) Feature of a Clinical 64-slice Cone-Beam Spiral CT Scanner.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4405954.html