The quantitative ability of PET/CT allows the widespread use in clinical research and cancer staging. However, metalartifacts induced by high-density metal objects degrade the quality of CT images. These artifacts also propagate to the corresponding PET image and cause a false increase of 18F-FDG uptake near the metal implants when the CT-basedattenuationcorrection (AC) is performed. In this study, we applied a model-basedmetalartifact reduction (MAR) algorithm to reduce the dark and bright streaks in the CT image and compared the differences between PET images with the general CT-based AC (G-AC) and the MAR-corrected-CT AC (MAR-AC). Results showed that the MAR algorithm effectively reduced the metalartifacts in the CT images of the ACR flangeless phantom and two clinical cases. The MAR-AC also removed the false-positive hot spot near the metal implants of the PET images. We conclude that the MAR-AC could be applied in clinical practice to improve the quantitative accuracy of PET images. Additionally, further use of PET/CT fusion images with metalartifactcorrection could be more valuable for diagnosis.
關聯:
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
648(Supplement 1):S142–S145
