Metal artefact reduction in gemstone spectral imaging dual-energy CT with and without metal artefact reduction software

Young Han Lee, Kwan Kyu Park, Ho Taek Song, Sungjun Kim, Jin Suck Suh

Research output: Contribution to journalArticle

142 Citations (Scopus)


Objective To assess the usefulness of gemstone spectral imaging (GSI) dual-energy CT (DECT) with/without metal artefact reduction software (MARs). Methods The DECTs were performed using fast kV-switching GSI between 80 and 140 kV. The CT data were retroreconstructed with/without MARs, by different displayed fields-of-view (DFOV), and with synthesised monochromatic energy in the range 40-140 keV. A phantom study of size and CT numbers was performed in a titanium plate and a stainless steel plate. A clinical study was performed in 26 patients with metallic hardware. All images were retrospectively reviewed in terms of the visualisation of periprosthetic regions and the severity of beam-hardening artefacts by using a five-point scale. Results The GSI-MARs reconstruction can markedly reduce the metal-related artefacts, and the image quality was affected by the prosthesis composition and DFOV. The spectral CT numbers of the prosthesis and periprosthetic regions showed different patterns on stainless steel and titanium plates. Conclusion Dual-energy CT with GSI-MARs can reduce metal-related artefacts and improve the delineation of the prosthesis and periprosthetic region. We should be cautious when using GSI-MARs because the image quality was affected by the prosthesis composition, energy (in keV) and DFOV. The metallic composition and size should be considered in metallic imaging with GSI-MARs reconstruction. Key Points • Metal-related artefacts can be troublesome on musculoskeletal computed tomography (CT). • Gemstone spectral imaging (GSI) with dual-energy CT (DECT) offers a novel solution • GSI and metallic artefact reduction software (GSI-MAR) can markedly reduce these artefacts. • However image quality is influenced by the prosthesis composition and other parameters. • We should be aware about potential overcorrection when using GSI-MARs.

Original languageEnglish
Pages (from-to)1331-1340
Number of pages10
JournalEuropean Radiology
Issue number6
Publication statusPublished - 2012 Jun 1


All Science Journal Classification (ASJC) codes

  • Radiology Nuclear Medicine and imaging

Cite this