Cost-effective Total-Body PET with Axial and Transverse Gaps

With significantly higher sensitivity compared to conventional PET, total-body PET offers improvements in image quality and quantification to enable faster scan times, lower dose, and new applications with novel tracers. However, the higher system cost hinders wide application of total-body PET technology in the clinic. Cost-effective total-body PET designs using fewer detectors in sparse configurations can be a potential solution. This paper aims to design sparse PET systems with both axial and transverse gaps, acquire and reconstruct patient images with no artifacts, and evaluate lesion quantification and detectability in these systems by leveraging the advantages of time-of-flight (TOF) technology. Taking PennPET Explorer as a template, two sparse configurations maintaining the same long axial field of view (LAFOV) of 142 cm were designed with the same total detector saving of 43% (around 1/3 of all line of response (LORs) remain). One design, called Sparse PET MT, includes more transverse gaps, while the second design with more axial gaps is called Sparse PET MA. We acquired patient data on the PennPET Explorer and generated data from Sparse PET MT and MA by discarding missing LORs. We embedded lesion data with varying uptake levels in lung and liver regions of measured patient FDG data. Images were analyzed using bias and precision metrics in measured lesion uptake as well using generalized scan statistics methodology to estimate lesion detectability. Our results show that the two sparse PET designs provide comparable lesion uptake measurement to the PennPET Explorer with no visual artifacts in a 1 min scan. Detailed evaluation will be performed for lesion quantitation and detectability measurements as a function of varying lesion uptake values.