Image Reconstruction Of Panel PET Based On Time Of Flight Information

Positron Emission Tomography (PET) , a kind of noninvasive imaging technique, can provide three dimensional tomographic images of radiotracer distribution and detect the metabolic changes, biochemical reactions and function of animal/human body's internal organs by noninvasive, quantitative and dynamic method.Panel PET employs panel detector structure. Comparing with the conventional PET of ring scanner, it presents several advantages such as lower cost, lower complexity of the installation, higher reliability, and the flexible structure which promote the development and application of panel PET. As the Time Of Flight (TOF) information brought in, the image quality of reconstruction in conventional ring PET can be improved, which has been carried out in the commercial PET. However, the effect of TOF information on image reconstruction of panel PET is unclear and needs further study.In this thesis, the influence of image quality of reconstruction which applied TOF information has been discussesd on the assumption that the time resolution of 30~100ps has been acquired. First, we set the simulation model of panel PET, select the proper source for the evaluation of the results and get the projection data for the reconstruction with TOF information. We use point source to test the spatial resolution, more complex source to test the quality of reconstruction images. Then we choose appropriate conventional reconstruction algorithm for panel PET and investigate the algorithm base on TOF information, analysis and evaluate the reconstruction results. According to the results, in the case of time resolution around less than 100ps, the spatial resolution is getting worse along with time resolution getting worse; but in the case of time resolution about more than 100ps, the spatial resolution is becoming an invariable value. Further more, the results indicate the quality of reconstruction image based on TOF information is superior conventional reconstruction image, if we can get time resolution that is good enough. Therefore, the thesis proves that we can get better quality of reconstruction image based on TOF information with good enough time resolution.