TY - JOUR T1 - Comparison of the Physical Performance of two Siemens Biograph Vision PET models with Extended Axial Field-Of-View Using Sparse Detector Blocks JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1156 LP - 1156 VL - 62 IS - supplement 1 AU - Nicolas Karakatsanis AU - Mohammad Nehmeh AU - Mercy Akerele AU - Girish Bal AU - Maurizio Conti AU - Sadek Nehmeh Y1 - 2021/05/01 UR - http://jnm.snmjournals.org/content/62/supplement_1/1156.abstract N2 - 1156Objectives: To evaluate and compare the physical performance of a simulated clinical PET scanner with extended axial field-of-view (AFOV) using sparse detector block rings and checkerboard detector block configurations respectively. Methods: The Biograph Vision (Siemens Healthineers) scanner adopts a compact configuration of 8 PET detector rings, each of 38 detector modules. Each detector module consists of 4 (transaxial) x 2 (axial) blocks of 5 x 5 LSO detector elements (3.2 x 3.2 x 20 mm3) coupled to a 4 x 2 SiPM array. The scanner AFOV was extended by adopting two sparse detector configurations: (i) sparse block ring (SBR), where all block rings were interleaved with axial gaps of 16 mm, and (ii) sparse checkerboard block (SCB), where all blocks were arranged in a checkerboard pattern and interleaved with 16mm gaps in both axial and transaxial directions. A limited continuous bed motion (limited-CBM) of 32 mm (corresponding to 2 x axial dimension of the detector block) along the axial direction was used to balance the axial sensitivity profile between the gaps and detectors. Thus, the AFOV was extended from 25.6 cm (compact) to 52.8 (SBR) and 54.4 cm (SCB) while preserving the total number of detector blocks and readout electronics. All configurations and limited-CBM acquisitions were simulated using the Monte Carlo GATE package. The physical performance (spatial resolution, sensitivity, and Image Quality) for the SBR and SCB were assessed per NEMA NU 2-2012 standards and compared to that of the original Biograph Vision model with compact detectors configuration (gold standard). The simulation scan time for the IEC phantom for each of the three scanner models was defined as per NEMA guidelines, thus resulting in 3 min for the compact and 9 min for the SBR and SCB models due to the differences in AFOV lengths. Image reconstruction was performed with STIR’s open-source OSEM algorithm (19 subsets, 1-5 full iterations) with normalization, attenuation, randoms, and scatter corrections. The contrast-to-noise ratio (CNR) of the 10 mm diameter sphere was measured to optimize the iterations number. Time-of-Flight and resolution recovery were not included. Results: All IEC phantom spheres (from 37 to 10 mm diameter) in the reconstructed PET images were detectable without artifacts. The percent contrast recovery (% CR), background variability (% BV), and CNR were comparable (mean difference within 4%) between SBR and SCB configurations. The % CR of the compact geometry for both 3 and 9 min scans was also comparable (mean differences within 4%) to that of sparse geometries. SBR and SCB Vision attained superior % BV and CNR scores by 10% on average compared to 3 min compact Vision scans. However, the % BV and CNR of the compact geometry was superior by 10% for matched scan times (9 min). SBR, SCB, and compact models exhibited system sensitivities of 16.6, 16.8, and 15.8 kcps/MBq respectively. The increased sensitivity of the sparse detector configurations may be due to the larger detector surface at the blocks side exposed to incident gammas entering through gaps. Compared to the compact model, both SBR and SCB exhibited less steep drop in sensitivity between the center and the edge of the AFOV. On average, SBR and SCB showed comparable spatial resolution in all directions, yet, up to 1.5 mm worse than those of the compact geometry, due to the larger acceptance angle of the longer AFOV sparse geometries. Conclusions: The extension by more than 100% of the AFOV of state-of-the-art clinical PET systems, such as Biograph Vision, without increasing the detector volume is feasible and can lead to a mildly higher system sensitivity and comparable image quality, yet at the cost of a slight reduction in spatial resolution due to the larger acceptance angle. SCB offered a mildly higher system sensitivity and image quality compared to SBR, yet the later may be more feasible to support the development of novel PET systems with adaptive AFOV. ER -