Low-dose Dual-energy CT Iterative Reconstruction And Robust PET Monto Carlo Simulation

Positron emission tomography(PET)is one of the most advanced technology in field of molecular imaging,which can reflect the metabolism distribution within human body.Computed tomography(CT)imaging can clearly provide anatomic structure information with high space-time resolution,which has been widely used in the clinic.There are two problems with single PET imaging system:(1)functional lesions cannot be anatomically located;(2)the attenuation correction can't be properly solved.The appearance of PET/CT imaging can solve the above two problems:(1)introducing anatomical location information by the fusion of CT images and PET images;(2)performing the attenuation correction of PET image by using the CT image.In recent years,PET/CT has been widely used to diagnose various diseases,for example,cancer,neuropathy and cardiovascular diseases.Attenuation correction is essential for PET images and can effectively improve the PET image quality.Dual-energy CT can obtain the conposed 511 ke V image by spectrum analysis,then caculate the attenuation factor at 511 keV,further achieve the purpose of accurate attenuation correction.For dual-energy CT,high levels of X-ray exposure may generate healthy risks,so it is anticipated that reducing radiation dose in DECT imaging is very important.However,reducing radiation dose may lead to excessive noise in the DECT images reconstructed by conventional algorithms.To our knowledge,there are two data acquisition schemes for low-dose CT image:reduce the X-ray flux towards each detector element for radiation dose reduction;and decrease the number of X-ray attenuation measurements across a whole object to be reconstructed.Statistical iterative reconstruction method is the most typical which can model the projection data and the image geometry by a cost function,and then minimizes the cost function to obtain high-quality images in case of low-dose.For PET?DECT,the DECT image quality can directly influence the composed image accuracy.To solve the abovebentional DECT imaging problem,we did the following works:(1)Iterative reconstruction for dual energy CT with an average image-induced nonlocal means regularization:Based on the success of edge-preserving non-local means(NLM)filtering in CT imaging and the intrinsic characteristics underlying DECT images,i.e.,global correlation and non-local similarity,a NLM-based regularization is incorporated into the penalized weighted least-squares(PWLS)framework.Specifically,the presented NLM-based regularization is designed by averaging the acquired DECT images,which takes the image similarity within the two energies into consideration.In addition,the garantee item takes into account the DECT data-dependent variance.Amplified experiments validate the feasibility and efficacy of the presented PWLS-aviNLM algorithm,including XCAT phantom study?physical phantom study and clinical experiments.(2)Dual-energy CT imaging with low-mAs and sparse-View combination scan protocol:In this study,a new data acquisition scheme was proposed for DECT imaging,incoprating the two low-dose acquisition methods for CT(low-mAs and sparse-view).Firstly,the low energy image acquired by low-mAs scheme is processed by a sinogram restoration algorithm to obtain high-quality images.Then,high-energy image data is scaned by sparse-view method.Based on the unique characteristics of dual-energy CT,we propose a new method of P WLS iterative reconstruction algorithm for the sparse-view high-energ image,which introduces a non-local prior from the restored low-energy image.As the current domestic medical equipments limited by the protection of trade secrets are difficult to open the original data,and to meet the data requirements of the researchers.To study the PET/DECT imaging system,based on the GATE platform,we designed two PET system simulated models,which are Inveon PET scanner and Biograph mCT scanner.GATE simulation platform based on GEANT4 can accurately simulate the photon's Compton effect,Rayleigh scattering and photoelectric effect,and can track the movement of photons in detail.At the same time,we validated the accuracy and feasibility of the two simulation models.The work on Monte Carlo simulation can be summarized as follows:(1)Validation of Inveon small animal PET system:The simulation results were validated against experimental or published data in accordance with the NEMA NU-4 2008 protocol for standardized evaluation of spatial resolution,sensitivity,scatter fraction(SF)and noise equivalent counting rate(NECR)of a preclinical PET system.An agreement of less than 18%was obtained between the radial,tangential and axial spatial resolutions of the simulated and experimental results.The simulated peak NECR of mouse-size phantom agreed with the experimental result,while for the rat-size phantom simulated value was higher than experimental result.The simulated and experimental SFs of mouse-and rat-size phantom both reached an agreement of less than 2%.It has been shown the feasibility of our GATE model to accurately simulate,within certain limits,all major performance characteristics of Inveon PET system.(2)Validation of Biograph mCT clinical PET system:Due to the limitation of the small animal PET system,we have further designed a clinical PET system simulation model.The design of the model is based on the GATE simulation platform.At the same time,we conduct a comprehensive verification and validation according to the widely used NEMA NU2-2007 standard protocol.The contents include three aspects:detector sensitivity,spatial resolution,scatter coefficient and counting rate.The experimental results show that the simulation model has a high degree of coincidence with the real Biograh mCT PET system,and can accurately reflect the characteristics of the system.Based on the simulation model,we can take some further performance for studying the process of PET system and PET imaging.PET simulation model is established for the further study of PET and PET?DECT imaging,which can offer the original projection data.