Daniel Prell Dipl Phys, Abstract Co-Author: Nothing to Disclose

Yiannis Kyriakou PhD, Abstract Co-Author: Nothing to Disclose

Michael Meyer, Abstract Co-Author: Nothing to Disclose

Marcel Beister, Presenter: Nothing to Disclose

Willi A. Kalender PhD, Abstract Co-Author: Consultant, Siemens AG Consultant, Bayer AG Founder, CT Imaging GmbH Scientific Advisor, CT Imaging GmbH Shareholder, CT Imaging GmbH Founder, Artemis Imaging GmbH CEO, Artemis Imaging GmbH Shareholder, Artemis Imaging GmbH

The aim of this study was to develop a physics-based metal artifact reduction (MAR) method for clinical CT without a priori knowledge.

Our approach uses a combination of two novel correction algorithms which aim at reducing the physical causes for metal artifacts. Firstly, a multi-dimensional directional filtering algorithm (MDF) reduces the metal induced noise artifacts by convolving the metal traces in the raw data with a sinusoidal smoothing function and attenuation-dependent combination weights. Secondly, an empirical beam hardening correction (EBHC) reduces beam hardening artifacts induced by metallic implants by using a material segmentation followed by a forward projection of each material. Combining the material raw data thus obtained and back projecting each sum yields correction images which are linearly combined with the initial CT image. The combination coefficients are determined by empirically minimizing the total variation of the obtained images. Performance was evaluated on phantom and patient datasets with uni- and bilateral hip endoprostheses for a clinical CT system (SOMATOM Definition, Siemens Healthcare, Germany). The fidelity of correction was investigated in subtraction images. Different regions of interest (ROIs) were evaluated in terms of CT value consistency and image noise. Results were compared to standard linear interpolation correction (LI).  

The physics-based approach achieved an appropriate metal artifact and noise reduction of up to 70% to the noise level of a scan w/o implants. New artifacts as introduced by LI did not appear. CT values in the vicinity to the implant were well restored without resolution losses (deviations to measurements without implants were reduced below 3 HU). Imaging of soft tissue structures, e.g. fat and muscle tissue, surrounding the implant and in greater distance was highly improved. Our method was always superior to the application of LI and offered a better detail preservation.

Our approach reduces the physical causes contributing to metal artifacts and significantly improves image quality.

Metal artifacts may prevent proper diagnosis. The reduction of obscuring metal artifacts can lead to an improvement of diagnostics.

Prell, D, Kyriakou, Y, Meyer, M, Beister, M, Kalender, W, Reducing Metal Artifacts in Computed Tomography Caused by Hip Endoprostheses Using a Physics-based Approach.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9008514.html