Abstract:1120 Objectives: PET human brain imaging has evolved dramatically, with the availability of a vast array of specific radiotracers, leading to numerous imaging paradigms to measure a wide range of brain targets and to assess neurotransmitter and receptor dynamics. Brain-dedicated PET systems could offer important advantages over currently available whole-body PET systems in terms of sensitivity and resolution. However, state-of-the-art for dedicated brain PET has not progressed beyond the HRRT, designed over 20 years ago. Therefore, there is a compelling need to build next generation human brain PET systems that incorporate significant advances in sensitivity and effective spatial resolution. This is the goal of the NeuroEXPLORER (NX) project. Methods: Experience with >4500 human brain HRRT studies at Yale has shown that sensitivity is a major limiting factor, i.e., radioactivity images are often noisy, and the best possible resolution cannot be achieved due to limited counts. Our experience with the total-body uEXPLORER system at UC Davis has shown that ultra-high sensitivity enables high SNR for short scans and parametric imaging at high spatial resolution with no smoothing. Thus, our primary NX design goals are three-fold. 1) Ultra-high sensitivity, to be achieved by maximizing the coincidence acceptance angle, with a long axial field-of-view (aFOV) together with excellent sensitivity-boosting time-of-flight (TOF). 2) To achieve exceptional image resolution through reduction in the size of detector elements, depth of interaction (DOI) readout, and corrections for inter-crystal scatter (ICS). Significant improvements in spatial resolution are needed to image small structures, reduce the partial volume effect, and measure the input function (IF) in the carotid arteries. 3) Head motion is a critical limit to image resolution and quantification accuracy. Thus, our third NX design goal is continuous motion tracking and correction. Results: The NX design is a cylinder with diameter and aFOV of ~50 cm. The system consists of 5 complete detector rings, each with 20 modules. An additional 6th ring has 3 blocks removed from each side to accommodate shoulders, to place the brain in the center of the aFOV. The 18-mm deep LYSO crystals have an in-plane dimension of 1.5 mm, leading to a simulated resolution of 1.6-1.8 mm. These are organized in micro-blocks of 8 crystals read out by 4 SiPMs with energy resolution 10-fold higher effective sensitivity than the HRRT for the brain, with an even greater advantage for the carotids. To provide even higher resolution (~ 1mm) to measure the carotid IF, the NX can support additional ultra-high resolution detector panels. Head motion tracking is performed with a real-time stereovision system using structured light. Initial comparison to the Vicra (which uses a tool mounted on the subject’s head) is underway with existing scanners. To optimize reconstruction and quantification accuracy, high-resolution brain simulations and novel phantom configurations are being designed. Conclusions: A next-generation ultra-high performance human brain PET, the NeuroEXPLORER, has been designed. Following manufacturing and testing, we will apply human imaging paradigms to demonstrate the effectiveness of the NX: 1) showing the dramatic sensitivity increase compared to the HRRT, 2) leveraging high sensitivity to reliably measure uptake in small nuclei, and 3) opening new frontiers of imaging neurotransmitter dynamics. The ultimate goal is a fully functioning and characterized system that dramatically expands the scope of brain PET protocols and applications. Research support: U01EB029811

Depth-of-interaction Reconstruction in NeuroEXPLORER

Improving the Off-Center Resolution in Position Emission Tomography Using Depth-of-interaction Information

Simulation study of a brain PET scanner using TOF-DOI detectors equipped with first interaction position detection.

Quantitative Accuracy Assessment of the NeuroEXPLORER for Diverse Imaging Applications: Moving Beyond Standard Evaluations

Development of a Depth of Interaction Detector for Γ-Rays

Impact of Depth-Of-Interaction on Image Resolution in Long Axial-Fov Pet

Spatial resolution recovery utilizing multi-ray tracing and graphic processing unit in PET image reconstruction.

Imaging performance evaluation in depth of interaction PET with a new method of sinogram generation: A Monte Carlo simulation study

Design of the NeuroEXPLORER, a Next-Generation Ultra-High Performance Human Brain PET Imager

Accurate Image Reconstruction for DOI-PET Systems and Its Implications for the Development of Economic, Compact PET (ezpet) Systems

The Effect of a Partial Shoulder PET Ring in the NeuroEXPLORER Design

Single-Ended Readout Depth-of-Interaction Measurements Based on Random Forest Algorithm

Virtual Cylindrical PET for Efficient DOI Image Reconstruction with Sub-millimetre Resolution

Effect of depth of interaction resolution on the performance of SIAT aPET

Accurate Image Reconstruction with Computed System Response Matrix for a High-Sensitivity Dual-Head PET Scanner

Improving depth-of-interaction resolution in pixellated PET detectors using neural networks

Pet Image Resolution Uniformity Improvements Using Deep Learning

Image Reconstruction with DOI Rebinning and PSF Modeling for the Quad-Head PET

Accurate image reconstruction using DOI information and its implications for the development of compact PET systems

A multi-resolution TOF-DOI detector for human brain dedicated PET scanner

Hybrid Model of Photon Propagation Based on the Analytical and Monte Carlo Methods for a Dual-Head PET System