Maximizing the Performance of Whole-Body PET Scanner - Summary The key performance characteristics of positron emission tomography (PET) are its sensitivity and spatial resolution. Higher sensitivity and spatial resolution allow for better diagnostic precision and patient outcomes. However, state-of-the-art whole-body PET scanners report spatial resolutions of ~3-5 mm at the center and ~5- 8 mm (FBP reconstruction) at a radial distance of 20 cm off-center of the scanner's field-of-view (FOV), with corresponding volumetric resolutions of ~25-90 mm3 and 55-250 mm3, respectively. This limited spatial resolution leads to significant limitations in using whole-body PET for applications that require high spatial resolution. Additionally, time-of-flight PET (TOF-PET) can provide an effective gain in sensitivity. Still, the coincidence time resolutions (CTRs) of state-of-the-art whole-body PET scanners, around 180-400 ps, can be further improved. We have worked over several years to develop next-generation whole-body PET scanners that can achieve theoretically achievable spatial resolution and sub-150 ps CTR. Our goal is to achieve sub-2 mm spatial and ~8 mm3 volumetric resolution (FBP reconstruction) across the entire FOV of scanners, which are ~2x and >6x better than those of the best available whole-body PET scanner, respectively, and comparable to state-of-the-art brain- dedicated PET scanners (e.g., NeuroEXPLORER). The spatial resolution can be further improved to better than 1.5 mm using OSEM reconstructions. This advancement will have profoundly transformative implications for the application of whole-body PET in clinical research and diagnostics that available scanners can not do or do not do well. For example, it will allow for 1) imaging heterogeneous dose distributions, transmural cardiac defects, the spinal cord, and quantitative imaging of vessel walls and lymph nodes, et al., 2) obtaining image-derived input function from the hepatic portal vein directly, and 3) obtaining high resolution images of organs such as the brain and the breast that are currently obtained using dedicated PET scanners. It could open new but currently unknown applications by enhancing the performance, as demonstrated by the EXPLORER PET scanner. We have optimized the design of the 16 x 16 LYSO arrays with a 1.6 mm pitch (for high spatial resolution) and 20 mm thickness (for high sensitivity). We have also developed and evaluated essential ASICs optimized for dual-ended readout detectors by working with our industrial partner, Canon Medical Research Institute USA, Inc., over the last three years. While our long-term goal is to ultimately develop a complete whole-body TOF-DOI PET scanner with a 1.63-meter axial FOV using our optimized detectors and electronics, in this application, we will construct and evaluate a proof-of-concept, one-detector ring whole-body TOF-DOI PET scanner with a 25.6 mm axial FOV, and compare image quality to that obtained using the state-of-the-art whole-body PET scanners and brain PET scanners. Furthermore, to guide future research on CTR enhancement, we will experimentally investigate the impact of further enhanced CTRs of 10, 25, 50, 75, and 100 ps on image quality using our developed activity painting digital phantom creation method and events acquired from the proposed PET scanner.
