3D list-mode surrogate-MLEM for dual-scanner PET image reconstruction

Positron Emission Tomography (PET) plays a crucial role in cancer diagnosis and monitoring despite its comparatively lower resolution to CT and MRI. Limited-angle, high-resolution PET scanners attempt to balance system cost with high-resolution. In this study, we propose a novel approach using a dual-system setup constituting a low-resolution 2π PET scanner and a high-resolution limited-angle PET scanner. We introduce a 3-D list-mode Maximum Likelihood Expectation Maximization (MLEM) algorithm, termed Surrogate MLEM, which incorporates data from non-synchronous imaging on both scanners. We perform a Monte Carlo simulation of this experimental setup using the high-resolution UCSC 2-Panel PET Scanner (2P) and the lower-resolution GE Discovery 690 WB PET Scanner, comparing the images from each system and the surrogate model. Our results demonstrate the Surrogate Scanner significantly reduces limite-dangle artifacts of the 2P Scanner and enhances contrast recovery over both systems. This approach shows promise for enhancing PET imaging in challenging anatomical regions like the head and neck of asynchronous coincidences.