A Method For Metal Artifacts Reduction Using Virtual Dual-energy CT Images Generated From Single Energy CT Scans

Metal artifacts caused by metal implants in computed tomography(CT)may seriously interfere with the results of clinical diagnosis and impact the accuracy of dose calculation in treatment planning for radiation therapy.Nowadays,conventional methods for metal artifacts reduction(MAR)mainly focus on processing the projection data before reconstructing a CT image data set,which cannot be applied in processing the reconstructed CT images.A new method to remove metal artifacts utilizing virtual dual-energy CT image sets generated from monoenergetic CT images and dual-energy CT subtraction is presented in this paper.CT image sets used in this paper were derived from Discovery CT 750HD(General Electric Company,USA).The head/neck protocol with tube voltage of 120 k Vp and slice thickness of 2.5 mm was used.Optimized conversion model from CT numbers to linear attenuation coefficients(LAC)was applied to calculate an accurate LAC map at target energy.According to mass attenuation coefficients(MAC)of base materials from the National Institute of Standards and Technology(NIST),a LAC map at target energy was obtained,and then sets of CT images was derived from the LAC maps,which is at different but a known energy.Then,dual-energy log subtraction was applied to remove metal artifacts.In the first step,CT image sets of the CIRS Model 062 electron density phantom were used to verify the conversion accuracy of the improved bilinear transformation model.It’s proved that,compared to GE dual-energy reconstruction images,the improved bilinear transformation model has a higher conversion accuracy for common tissues in human body,such as fat,soft tissue,muscle tissue and dense bone.Secondly,14 groups of clinical images were used to verify that the dual-energy log reduction has a significant effect for the removal of metal artifacts.Comparison between dual-energy log subtraction and polar coordinate transformation method was made to evaluate the effect of the metal artifacts reduction.