Combination of dual-energy computed tomography and iterative metal artefact reduction to increase general quality of imaging for radiotherapy patients with high dense materials. Phantom study. (September 2020)

• Record Type: Journal Article
• Authors: Pawałowski, Bartosz; Panek, Rafał; Szweda, Hubert; Piotrowski, Tomasz
• Is Part Of: Physica medica. Volume 77(2020)
• Page Start: 92
• Page End: 99
• Publication Date: 2020-09
• Subjects: Metal artefact reduction -- Dual energy CT -- iMAR -- Quality of CT images -- Imaging for radiation therapy; Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2020.08.009
• Languages: English
• ISSNs: 1120-1797

Abstract

Highlights: An overall improvement in image quality was obtained for the 70 keV PMR. iMAR method provided best reduction of metal artefacts observed on CT. Combining the 70 keV PMR with the iMAR bring benefits for CT use in RT. Abstract: Purpose: To evaluate the use of pseudo-monoenergetic reconstructions (PMR) from dual-energy computed tomography, combined with the iterative metal artefact reduction (iMAR) method. Methods: Pseudo-monoenergetic CT images were obtained using the dual-energy mode on the Siemens Somatom Definition AS scanner. A range of PMR combinations (70–130 keV) were used with and without iMAR. A Virtual Water™ phantom was used for quantitative assessment of error in the presence of high density materials: titanium, alloys 330 and 600. The absolute values of CT number differences (AD) and normalised standard deviations (NSD) were calculated for different phantom positions. Image quality was assessed using an anthropomorphic pelvic phantom with an embedded hip prosthesis. Image quality was scored blindly by five observers. Results: AD and NSD values revealed differences in CT number errors between tested sets. AD and NSD were reduced in the vicinity of metal for images with iMAR (p < 0.001 for AD/NSD). For ROIs away from metal, with and without iMAR, 70 keV PMR and pCT AD values were lower than for the other reconstructions (p = 0.039). Similarly, iMAR NSD values measured away from metal were lower for 130 keV and 70 keV PMR (p = 0.002). Image quality scores were higher for 70 keV and 130 keV PMR with iMAR (p = 0.034). Conclusion: The use of 70 keV PMR with iMAR allows for significant metal artefact reduction and low CT number errors observed in the vicinity of dense materials. It is therefore an attractive alternative to high keV imaging when imaging patients with metallic implants, especially in the context of radiotherapy planning.