Wouter J.H. Veldkamp PhD, Presenter: Nothing to Disclose

Raoul M. S. Joemai BSC, Abstract Co-Author: Nothing to Disclose

Koos Geleijns PhD, Abstract Co-Author: Nothing to Disclose

Metal shoulder or hip prostheses cause artifacts in CT images. The purpose of this work was to design a potentially fast and effective artifact suppression scheme directly applied to sinograms, to improve CT image quality.

Suppression of metal is performed directly in sinograms, making it potentially faster than current methods based on reconstructed images and forward-projection. Moreover, it enables correction for metal objects outside the field of view. Suppression involves two steps: metal prosthesis segmentation (1) is achieved by using a Markov Random Field Model (MRF). The segmented region is replaced (2) by interpolation. To preserve the structure of adjacent projections, corresponding pixel sites are determined at the different sides of the prosthesis based on different features. Interpolation is performed between the corresponding pixel sites. Adapted sinograms are back-projected to obtain reconstructed images. To provide a proof of concept, CT data of five patients with metal prosthesis were included in the study as well as CT data of a PMMA phantom with inserts (Teflon, PVC, Titanium). Qualitative improvement was assessed by comparing clinical images with artifact suppressed images. Quantitative accuracy was determined by comparing mean CT values and standard deviation (SD) values in phantom images with Titanium (original and suppressed) and without Titanium insert.

Artefacts in five patients were automatically suppressed with improved visibility of structures. From quantitative analyses it appeared that mean +/- SD values for Teflon with Titanium insert were 885 +/- 91 versus 895 +/- 49 with suppression (values without Titanium insert: 900 +/- 47). Values for 2 PVC inserts were on average -60 +/- 75 with Titanium insert. With suppression the values were -49 +/- 33 (values without Titanium insert: -62 +/- 35).

The automated artifact suppression design substantially reduces artifacts in clinical images. Phantom analysis showed reduced SD values close to the SD values without Titanium insert and thereby improved image quality. The CT value for Teflon (simulating bone) in suppressed images was similar to normal. For PVC (low contrast) the values were slightly higher than normal.

Metal artifacts in CT images can be successfully suppressed by adapting the original raw data, improving CT image quality for patients with metal hip or shoulder prostheses.

Veldkamp, W, Joemai, R, Geleijns, K, Automated Segmentation and Interpolation in Sinograms for Metal Artifact Suppression in CT.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8010847.html