Artifacts Caused by Breast Tissue Markers in a Dedicated Cone-beam Breast CT in Comparison to Full-field Digital Mammography.

RATIONALE AND OBJECTIVES: The purpose of this ex vivo study was to investigate artifacts in a cone-beam breast computed tomography (CBBCT) caused by breast tissue markers.

MATERIALS AND METHODS: Breast phantoms with self-made tissue pork mincemeat were created. Twenty-nine different, commercially available markers with varying marker size, composition, and shape were evaluated. A dedicated CBBCT evaluation of all phantoms was performed with 49 kVp, 50 and 100 mA, and marker orientation parallel and orthogonal to the scan direction. The resultant images were evaluated in sagittal, axial, and coronal view with a slice thickness of 0.5 mm. Additionally, measurements of all markers in the same directions were done with full-field digital mammography.

RESULTS: All markers were visible in full-field digital mammography without any artifacts. However, all markers caused artifacts on a CBBCT. Artifacts were measured as the length of the resulting streakings. Median length of artifacts was 7.2 mm with a wide range from 0 to 48.3 mm (interquartile range 4.3-11.4 mm) dependent on composition, size, shape, weight, and orientation of the markers. The largest artifacts occurred in axial view with a median size of 12.6 mm, with a range from 0 to 48.3 mm, resulting in a relative artifact length (quotient artifact in mm/real physical length of the marker itself) of 4.1 (interquartile range 2.3-6.1, range 0-8.7).

CONCLUSIONS: Artifacts caused by markers can significantly influence image quality in a CBBCT, thus limiting primary diagnostics and follow-up in breast cancer. The size of the artifacts depends on the marker characteristics, orientation, and the image plane of reconstruction.