Investigation of a high-resolution detector for whole-body PET imaging

Increased focus towards improved detector spatial resolution in PET (for whole-body as well as small animal imaging) has led to the use of smaller crystals in some form of light sharing detector design. We have performed simulations for a generic whole-body scanner with varying crystals sizes in order to better understand the limitations in achieving good uniform spatial resolution using discrete crystal based detectors in a whole-body scanner. These results show that using small 2-mm crystals leads to a noticeable improvement in the reconstructed spatial resolution of whole-body scanners due to better sampling in the sinogram. Over the years we have investigated using large continuous detectors with Anger positioning, cutting slots in continuous detectors to produce semi-continuous detectors, and more recently pixelated Anger-logic detectors. Our aim is to develop a new detector concept that incorporates high detector sampling with very good intrinsic spatial resolution to produce high resolution reconstructed images. In this work we investigate three different techniques for achieving our goals: (1) Using regions-of-interest (ROI) which are half the crystal size in the crystal identification map. (2) Dual layers of 4×4×15-mm³ offset crystals. (3) Large continuous crystals with 3D positioning of interaction (3D crystals) using maximum likelihood positioning algorithm. Our results show a reasonable improvement in spatial resolution (1-mm at center) using smaller ROI´s and a similar improvement in off-center positions using the offset crystals. A more noticeable improvement is observed with the 3D crystal detector where the resolution seems to be intrinsically limited by the physics of positron annihilation and amount of Compton scatter in the crystal.