| Nuclear accidents often lead to the inhalation of a variety of nuclides,resulting in lung internal contamination,usually with low activity.Although parallel beam collimator imaging has been widely used in medical gamma camera,and has the advantages of high spatial resolution and good image quality,but its detection efficiency is very low,so it is difficult to image for internal contamination in lungs.Therefore,it is of great significance to find out the distribution of radionuclides in vivo through the study of multi nuclide internal pollution energy spectrum imaging measurement method and technology,so as to reduce the measurement error of internal pollution and improve the accuracy of internal radiation dose assessment.In this paper,the distribution of radionuclides in vivo is inversed by MPA-MURA coding.We use Monte Carlo method to simulate the MPA-MURA coding imaging of internal contamination in lungs.Three distributions of four different nuclides in the lung were simulated in GEANT4?geometry and tracking?and their MURA coded imaging was performed.In the same case,the parallel beam collimator imaging with the same spatial resolution in design is compared.Finally,based on the operability of the project,the"reduced"MPA-MURA coding imaging is simulated.The main research is as follows:1)Collimator design.The design process of MPA-MURA encoded hole collimator is described in detail in this article.Based on the size of both lungs in the thoracic model,MPA-MURA suitable for intra-pulmonary contamination imaging was designed with a specification of 301×301,material of tungsten,and an open basic unit of 1mm×1mm square hole with a size of 301mm×301mm×4mm,which allows accurate imaging of lung radiation sources.A parallel beam?PB?collimator model of the same resolution with an opening interval d of 1 mm,a shield thickness t of 2 mm,and an opening depth l of 80 mm was also designed,and the final consideration was engineering feasibility to reduce the size of the detector to 1?3 of the original size with the original field of view unchanged,and to design an intra-pulmonary contamination imaging system on this basis.Its specification is 157×157,the material is tungsten,the basic unit of the opening is 0.95mm×0.95mm square hole,and the size is150.72mm×150.72mm×4mm.2)The?-decoding algorithm,fine balance decoding algorithm?FSBD?,and maximum likelihood estimation?MLEM,Maximum Likelihood Estimation Method?reconstruction algorithm for the image of the internal contamination distribution were investigated in the paper.The reconstructed images of the MLEM algorithm were 8.3%and 7.5%higher than the reconstructed images of the other two algorithms,respectively,in the case of the"circular"source of 131I.In this paper,the MLEM decoding algorithm was finally selected as the decoding algorithm for intra-pulmonary contamination imaging,and the inversion image was successfully obtained by calculating the nucleophile distribution image in vivo.3)The"zeroing"of the encoded image is a periodic extension of the"zeroing"around the encoded image obtained by the detector.The reconstructed image after"zeroing"improved resolution by 4.4 percent and signal-to-noise ratio by 14 times.In this paper,the mechanism of near-field artifact generation is analyzed,and in this paper,complementary imaging is used to significantly reduce the near-field artifacts and increase the image signal-to-noise ratio by 254.17%.4)This paper compares MPA-MURA imaging with parallel-hole imaging studies,where MPA-MURA collimator imaging achieves a spatial resolution of 2.51 mm for a131I?365ke V?point source and a signal-to-noise ratio of 3.98 d B for a 131I simplified bronchial source with a detection efficiency of1.69×10-3.The spatial resolution of MPA-MURA encoded imaging is less than that of parallel-beam collimator imaging,but its detection efficiency is two orders of magnitude higher.In this paper,a simulation study was conducted based on the scaled-down imaging of MPA-MURA imaging system,which obtained a lower signal-to-noise ratio and lower detection efficiency due to the larger object distance of the image. |
PDF Full Text Request