No abstract

In this article, we present the geometrical design and preliminary results of a high sensitivity organ-specific Positron Emission Tomography (PET) system dedicated to the study of the human brain. The system, called 4D-PET, will allow accurate imaging of brain studies due to its expected high sensitivity, high 3D spatial resolution and, by including precise photon time of flight (TOF) information, a boosted signal-to-noise ratio (SNR). The 4D-PET system incorporates an innovative detector design based on crystal slabs (semi-monolithic) that enables accurate 3D photon impact positioning (including photon Depth of Interaction (DOI) measurement), while providing a precise determination of the photon arrival time to the detector. The detector includes a novel readout system that reduces the number of detector signals in a ratio of 4:1 thus, alleviating complexity and cost. The analog output signals are fed to the TOFPET2 ASIC (PETsys) for scalability purposes. The present manuscript reports the evaluation of the 4D-PET detector, achieving best values 3D resolution values of <1.6 mm (pixelated axis), 2.7±0.5 mm (monolithic axis) and 3.4±1.1 (DOI axis) mm; 359 ± 7 ps coincidence time resolution (CTR); 10.2±1.5 % energy resolution; and sensitivity of 16.2% at the center of the scanner (simulated). Moreover, a comprehensive description of the 4D-PET architecture (that includes 320 detectors), some pictures of its mechanical assembly, and simulations on the expected image quality are provided.

No abstract

In this work, we present the performance characteristics of the MAMMOCARE PET prototype based on an adaptation of the NU 4-2008 NEMA standard. MAMMOCARE is a project under the European Commission's 7th Framework programme to develop a breast biopsy system guided by a dedicated breast PET (dbPET) images. The PET system is formed by two rings with twelve detector modules each. The transaxial FOV is 170 mm and the axial FOV is 94 mm. The system can separate the detectors up to 60 mm in transaxial plane to allow the biopsy needle entrance.The acquisitions are reconstructed using the LMOS algorithm with tube-of-response (TOR) backprojector, 1 iteration and 16 subsets. The voxel and pixel sizes are (1 × 1 × 1) mm 3 and (1.6 × 1.6) mm 2 respectively. The radial resolution measured is 1.62 mm in the center of the FOV and 3.45 mm at 50 mm off the center in the radial direction using the closed configuration. In the open configuration the resolution reaches 1.85 mm and 3.65 mm at center and at 50 mm offcenter. The adapted recovery coefficients (ARC) are measured for six hot rods inside a cylindrical phantom with a warm background. The ratio between hot and background regions is 10. The ARC values for the closed configuration are 0.32, 0.77 and 0.96 for the inserts with a diameter of 4.5 mm, 8.3 mm and 25 mm, respectively. These values decrease to 0.16, 0.52 and 0.77 for the open configuration. The sensitivity measured using an energy window of 250 keV-750 keV is 3.6% and 2.5% for the closed and open configurations respectively. The NEC peak is 141 kcps@68 MBq and 147 kcps@78 MBq for closed and open configurations. The performance characteristics measured with the open ring configuration decreases with respect the closed configuration, however the values remain comparable to other dbPETs.

No abstract