Abstract
242468
Introduction: Over the past decade, there has been significantly increased interest in studying neurological disorder incidence and mortality across the globe [1], leading to a more rigorous requirement for brain-dedicated brain imaging systems with high sensitivity and high spatial resolution. Recent developments of the brain-dedicated Single Photo Emission Computed Tomography (SPECT) systems have offered a spatial resolution of 2.5-10 mm [2][3], however, due to the microscopic anatomy of the brain structures, a significant improvement is still required for the SPECT systems to better understand the physiology of several neurodegenerative disorders including Epilepsy, Alzheimer’s disease, and so on. We are currently developing a so-called Neuro-Scope system, which is a dedicated clinical brain scanner designed to offer a sub-mm or even sub-0.5 mm spatial resolution for imaging the micro-structures deep in the brain. In this paper, we present a design study and hardware design features for the Neuro-Scope system based on state-of-art CZT imaging spectrometers with an excellent energy resolution and a dual field of view (FOV) system gantry allowing the switching between stationary full-brain imaging and zoomed in the scan of local regions at an ultrahigh spatial resolution. The microscopic imaging mode relies on a compound-eye camera design of non-conventional micro-ring apertures to ensure an ultrahigh spatial resolution that offers a reasonable spatial resolution. [4].
Methods: The current design of the Neuro-Scope system is shown in Fig. 1A-B, 20 detector panels are arranged in a cylindrical geometry with each detector panel comprising of 16 high-performance CZT detectors [5] arranged in an 8 x 2 configuration for a total of 320 detectors (Fig. 1D). A single CZT crystal is 20 mm x 20 mm x 2 mm with 80 x 80 pixels of 250 µm pitch. Each detector panel is coupled to a collimator panel comprising of two-sets of aperture designs – pinhole and micro-ring. The knife-edge pinhole aperture is 2-mm in diameter focusing on a 20-cm diameter FOV. The micro-ring aperture (Fig. 1C) comprises of 2 pieces: an external part of 10-mm diameter opening (ro), which modulates the expected sensitivity of the aperture, and an internal part of a maximum diameter of 9.7-mm (ri) with a minification factor of 1:3 for a 5-cm diameter FOV. The difference between the outer and inner part radius defines the width of the ring (w = ro – ri) which governs the expected spatial resolution. Thus, a larger ‘ro’ and a smaller ‘w’ provides an aperture design with an ultra-high spatial resolution and high-sensitivity optimum for the development of a brain-dedicated SPECT system.
Results: The preliminary Monte Carlo simulations of the proposed system design with the pinhole apertures yields a peak sensitivity of ~0.13% with a >0.1% sensitivity for a central 18-cm diameter region (Fig. 2A) and a 6-mm spatial resolution. In the micro-ring aperture design, a peak sensitivity of ~0.19% is observed with an excellent spatial resolution of 1-mm (S/B 1:2 Fig. 2B). Additionally, as shown in Fig. 2B, we also simulate an ictal human brain perfusion phantom in mesial temporal lobe epilepsy (MTLE) [6] for both the aperture designs. The reconstructed image in the pinhole geometry (Fig. 2C), shows the entire brain phantom whereas in the micro-ring design we zoom into a smaller region around the temporal pole in the ictal hemisphere (smallest anatomical volume in the phantom).
Conclusions: In this work, we have demonstrated the preliminary performance of our high-sensitivity and ultra-high spatial resolution (1-mm) brain-SPECT system using Monte Carlo simulations. Further simulation and experimental studies will be performed to optimize the system design for the dual-FOV imaging along with exploration of the optimal aperture designs to achieve sub-mm resolution. Additionally, we will be investigating different acquisition scanning schemes to improve the angular sampling of system and enhance the spatial resolution.