ECT Image Reconstruction: A State Space Framework

Emission computed tomography (ECT) is the medical imaging technique that provides functional information of physiological activity by displaying the concentration distribution of radioisotope labeled chemical compounds or biological molecular which pre-injected into the human body before imaging process.As typical ECT modalities, positron emission tomography (PET) and single photon emission computed tomography (SPECT) have recently enjoyed increasingly widespread use in oncology, brain and neuro science, drug development and metabolic activity research.In most of application of ECT image, especially for tumor diagnostic early or staging, brain function and drug evaluation, both image quality and quantitative accuracy are important performance, which due to accurate and proper imaging reconstruction process. However, current statistical iterative reconstruction algorithms still could not deal with the complex statistical properties of ECT measurement data and noise and there still has been great difficulty in attenuation correction for SPECT image so that it could not be used for quantitative practice. On the other hand, uncertainty in system matrix still could not be modeled with current system response model, which greatly affects the quality of the reconstructed images.For real practice, there is still no unified standard the system evaluation and correction for small animal imaging system, which is also very important for the final image quality. And as a necessary basis for imaging system study, a experiment platform should be developed for the systematic reconstruction research.In this paper, we explore a reconstruction framework for ECT image based on state space principles. The proposed strategy formulates coherently treats the uncertainties of the statistical model of the imaging system and the noisy nature of measurement data. And as its application, we combines the both the radioactivity and attenuation estimation in one framework by a augmented state space model, then realized a joint estimation algorithm to solve the attenuation correction problem in ECT imaging.As another extension or the state space framework, we propose an algorithm for PET image reconstruction for the real world case where the PET system model is subject to uncertainties. The method is based on the formulation of PET reconstruction as a regularization problem and the image estimation is achieved with the aid of an uncertainty-weighted least squares framework.Besides the reconstruction algorithm framework establishing, this paper also try to improve the image quality and the whole EIT system performance by some system calibration process, and a system evaluation scheme together with the phantom design and data analysis procedure has been established and put into practical experiments for the first microPET scanner in China mainland.As another important issue for reconstruction and imaging system study, a high performance positron detector system is applied for the imaging experiment platform design, which collaborated with Hamamtsu Photoics K.K.. And beyond the platform implementing, a initial result for the dedicate system applied for positron planar imaging will be given in relative chapter.All the algorithm in this paper are evaluated with both digital phantom data and real phantom scanning data by some clinical PET or SPECT scanners in the second affiliate hospital of Zhejiang University Medical College.