LOR-based reconstruction for super-resolved 3D PET Images

Abstract

Positron emission tomography (PET) images usually suffer from incorrect quantification of the radioactive uptake of lesions due to low spatial resolution. To improve the spatial resolution based on wobble scanning, we previously proposed a sinogram-based super-resolution (SR) algorithm using a space-variant blur matrix. However, the algorithm may cause unwanted resolution loss owing to an inevitable interpolation process for preparing evenly spaced projections. In this dissertation, we propose a novel one-step line of response (LOR)-based SR framework for 3D PET images. In the framework, we efficiently determine a large number of space-variant point spread functions (PSFs) in the image space by using the scanner symmetries and the proposed PSF interpolation scheme based on non-rigid registration. Furthermore, to minimize the resolution degradation in the evenly spaced parallel-beam re-binning and to reduce the computational time in the graphics processing unit (GPU) implementation, we introduce parallel-friendly LOR reconstruction based on cone-beam reordering. We then obtain high resolution images via a one-step super-resolved 3D PET image reconstruction with low resolution sinograms measured through wobble scanning. Meanwhile, the wobble scanning requires a moving mechanism of the patient bed or a system gantry. To alleviate this problem, we propose to use respiratory motion rather than wobble motion for 3D PET SR of images. In this proposed framework, gated list-mode PET data are acquired in a free breathing condition as in the conventional protocol of PET imaging. In addition, we acquire two low-dose CT images in a breath-hold manner at exhale and inhale phases. Using the two low-dose CT images, we estimate the 4D motion vector field (MVF) and correspondingly generate a virtual 4D CT image. The CT images have much better spatial resolution than PET images and therefore the estimated MVF can be considered reliable for PET SR reconstruction. We then estimate space-variant PSFs in the imaging field of view using a minimum number of PSFs obtained through Monte-Carlo simulations. Finally, SR reconstruction is performed by incorporating the estimated MVF and space-variant PSFs. In the SR reconstruction, to minimize the resolution degradation in the evenly spaced parallel-beam re-binning and to reduce the computational time on the GPU, we introduce a parallel-friendly spanned line of response reconstruction technique based on fan-beam reordering. The two proposed frameworks commonly provide noticeable improvement on the spatial resolution of PET images with a considerable reduction of computational time.