3D PET image reconstruction based on the maximum likelihood estimation method (MLEM) algorithm

Abstract A positron emission tomography (PET) scan does not measure an image directly. Instead, a PET scan measures a sinogram at the boundary of the field-of-view that consists of measurements of the sums of all the counts along the lines connecting the two detectors. Because there is a multitude of detectors built in a typical PET structure, there are many possible detector pairs that pertain to the measurement. The problem is how to turn this measurement into an image (this is called imaging). Significant improvement in PET image quality was achieved with the introduction of iterative reconstruction techniques. This was realized approximately 20 years ago (with the advent of new powerful computing processors). However, three-dimensional imaging still remains a challenge. The purpose of the image reconstruction algorithm is to process this imperfect count data for a large number (many millions) of lines of response and millions of detected photons to produce an image showing the distribution of the labeled molecules in space.