TY - JOUR T1 - Design and Development of an MRI Compatible Human Brain PET Scanner by Using Dual-ended Readout Detectors JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1708 LP - 1708 VL - 62 IS - supplement 1 AU - Zhonghua Kuang AU - Ziru Sang AU - Xiaohui Wang AU - Dongfang Gao AU - Ning Ren AU - San Wu AU - Tianyi Zeng AU - Ming Niu AU - Longhan Cong AU - Zheng Liu AU - Tao Sun AU - Zhanli Hu AU - Junwei Du AU - Yongfeng Yang Y1 - 2021/05/01 UR - http://jnm.snmjournals.org/content/62/supplement_1/1708.abstract N2 - 1708Objectives: Dual-modality PET/MRI imaging is a very powerful tool for brain imaging since PET and MRI are important brain imaging tools and highly complementary. Dedicated brain PET scanner can provide much higher sensitivity and spatial resolution that are required for neuroimaging as compared to whole-body PET scanner. The aim of this work is to develop an MRI compatible human brain PET insert with a uniform high spatial resolution of ~1.5 mm within the whole field-of-view (FOV) and a peak sensitivity of higher than 10% by using dual-ended readout depth encoding detectors. Materials and Methods: The PET scanner has 224 detector modules arranged in 8 detector rings with 28 detectors per ring. The detector ring diameter of the scanner is 376.8 mm and the axial FOV is 329 mm. The system has an aperture of about 360 mm to allocate a dedicated brain birdcage RF coil. An MRI RF coil with 1 Tx-end and 47 Rx-end channels is designed for this PET insert. This defines a geometry with an axial FOV of 330 mm and a transaxial FOV of about 240 mm. Water cooling with aluminum nitride ceramics sheets and thermal conductivity copper tubes was used to ensure a stable operating temperature for the scanner. MRI shielding consists of carbon fiber tube and copper foil is used. The LYSO arrays have 26×26 crystals of 1.4 mm×1.4 mm×20 mm. BaSO4 with 0.08 mm thickness is used as the reflector between the crystals. SiPMs (Hamamatsu S14160-3050HS) with a size of 3.4 mm×3.4 mm and sensitive area of 3.0 mm×3.0 mm are soldered on readout PCB, forming a 10×10 array with a pitch size of 4.0 mm. The LYSO arrays are read out by two 10×10 SiPM arrays placed at the opposite ends. The PET insert uses 151424 crystal elements and 44800 SiPM pixels. The 100 pixels of a SiPM array are read out with a signal multiplexing readout circuit to reduce the number of signals from 100 to 4. The 8 output signals of one detector module are sent to a pre-amplifier board to be amplified and shaped for the energy measurement and the event timing is pick-off with a leading-edge discriminator by using the summing of the 8 signals. The energy and timing signals are then sent to the singles processing unit (SPU) by a long micro-coaxial cable. The system electronics consists of 28 SPUs with each processing 64 channel signals of 8 dual-ended readout detectors, a coincidence processing unit (CPU), a system clock and synchronizing board and power supply boards. The performance of the detector was measured by using both NIM and the system electronics. Results: All but the edge crystals can be clearly resolved from the flood histogram and the detectors achieve a DOI resolution of 2.33±0.26 mm, an energy resolution of 16.1±2.3% and a timing resolution of 1.60±0.18 ns. The MRI compatibility measurement of the PET detector shows that MRI imaging has negligible effect on the performance of PET detector and PET detector only has small effect on the signal-to-noise ratio of MRI image. Conclusions: An MRI compatible human brain PET insert consisting of 224 dual-ended readout detector modules is under development. The PET insert is expected to achieve a peak sensitivity of higher than 10% and a uniform spatial resolution of 1.5 mm. The manufacture of all components of the scanner are almost completed and the integration of the scanner is ongoing. An OSEM reconstruction algorithm is also under development for the scanner with a huge sinogram size of 358×400×49284 and an image size of 530×530×443. Acknowledgments: This work is supported by the Scientific Instrument Innovation Team of Chinese Academy of Sciences (GJJSTD20180002), the Chinese Academy of Sciences Engineering Laboratory for Medical Imaging Technology and Equipment (KFJ-PTXM-012) and the Hundred-Talent Program of the Chinese Academy of Sciences (Yongfeng Yang). Corresponding author: Yongfeng Yang, yf.yang@siat.ac.cn ER -