%0 Journal Article %A Mehrsima Abdoli %A Mohammad Ay %A Johan de Jong %A Jan Pruim %A Rudi Dierckx %A Habib Zaidi %T Reduction of artifacts arising from metal hip implants during CT-based attenuation correction in oncological PET-CT studies %D 2011 %J Journal of Nuclear Medicine %P 1990-1990 %V 52 %N supplement 1 %X 1990 Objectives Beam hardening artifacts caused by metal hip implants in CT images may erroneously increase or decrease the radiotracer’s measured uptake following CT-based attenuation correction of PET data. The purpose of this study is to assess the performance of our recently proposed metal artifact reduction (MAR) approach, which already proved to be efficient in reduction of dental filling artifacts, in metal hip implants. Methods Our proposed method substitutes the affected projection bins in the virtual sinogram space by a weighted sum of three values: 1) the original value, 2) one obtained by spline interpolation, and 3) values of the sinogram bins in the neighboring column of the sinogram matrix. This method was applied to 20 clinical PET/CT studies with metallic hip implants. The results were analyzed using Bland-Altman plots, paired t-test and another method allowing statistical analysis in the absence of gold standard called regression without truth (RWT) Data analysis was performed by defining 45 ROIs of the same size on each dataset. Results The Bland-Altman analysis proves the efficiency of the proposed MAR algorithm in reducing streak artifacts. The statistical analysis revealed statistically significant difference between the uptake values in regions with increased/decreased uptake and the corresponding regions in the corrected PET images (P-values<0.0001), whereas there is no significant difference in the unaffected regions (P-value=0.09). The RWT analysis shows similarity of the corrected data to the hypothesized gold standard. Conclusions The proposed MAR method reduces artifacts arising from metallic hip implants and improves the diagnostic quality and quantitative accuracy of corresponding clinical PET/CT images %U