Development of a Low Cost, Small Animal PET System

描述（由申请人提供）：提议的努力将开发一种低成本，超高分辨率，小动物PET成像系统。该提案R21阶段的关键组成部分是连续微型晶体元件（cMiCE）探测器设计的开发和可行性测试。目标是一个探测器的固有空间分辨率，将转化为相机图像分辨率约为1毫米全宽的一半最大（FWHM）和一个有效的探测器成像面积大于或等于使用相同光电倍增管（PMT）的离散晶体探测器的成像面积。使用1x1毫米或更小横截面晶体的超高分辨率探测器的成本是由切割和表面处理单个晶体以及手工组装模块的成本驱动的。通过使用单个25x25mm横截面晶体，与使用等效的1x1 mm晶体阵列相比，将实现重大的成本降低。连续型晶体探测器还具有更好的填充率和取样特性。小面积连续晶体实现的主要限制是由于边缘效应减少了有效成像面积。为了扩大晶体的成像范围，提出了一种基于统计的定位技术。第二个缺点是固有空间分辨率随着晶体厚度的增加而变宽。提高厚连续晶体探测器空间分辨率特性的技术也将被研究。探测器模块将使用目前可用的闪烁体材料和pmt；然而，该设计将很容易适应新技术，例如，雪崩光电二极管阵列，如果它们变得具有成本效益。将对探测器的性能进行仔细的表征。问题，如探测器的稳定性，探测器校准，计数率性能将包括在内。该提案的R33阶段将包括使用cMiCE探测器设计建立一个专用的小动物PET系统。优选的系统设计将具有探测器环直径约12厘米和轴向视场接近10厘米。蒙特卡罗模拟将用于优化估计任务性能的分辨率灵敏度（即对比度与噪声）权衡。将进行额外的模拟，以量化正电子范围对不同放射性同位素的影响。将研究有和没有分辨率恢复重建图像的定量精度（例如，正电子范围，光子共线性和探测器响应）。最后，研究了利用LRF的变化来估计厚晶设计的相互作用深度的可行性。

DESCRIPTION (provided by applicant): The proposed effort will develop a low cost, ultra-high resolution, small animal PET imaging system. The key component of the R21 phase of this proposal is the development and feasibility testing of a continuous miniature crystal element (cMiCE) detector design. The goals are a detector intrinsic spatial resolution that will translate into a camera image resolution of about 1 mm full width at half maximum (FWHM) and an effective detector imaging area greater than or equal to the imaging area of a discrete crystal detector using the same photomultiplier tube (PMT). The cost of ultra-high resolution detectors using 1x1 mm or smaller cross-section crystals is driven by the cost to cut and surface treat the individual crystals and to hand assemble the modules. By using a single 25x25 mm cross-section crystal, major cost reductions will be realized versus using an equivalent array of 1x1 mm crystals. A continuous crystal detector will also have better packing fraction and sampling characteristics. The main limitation to a small area continuous crystal implementation has been a reduced effective imaging area due to edge effects. A statistically based positioning technique is proposed to extend the imaging area of the crystal. A second drawback is that intrinsic spatial resolution tends to broaden as the crystal thickness increases. Techniques to improve the spatial resolution characteristics of thick continuous crystal detectors will also be investigated. Detector modules will use currently available scintillator materials and PMTs; however, the design will be easily adaptable to new technologies, for example, avalanche photodiode arrays if they become cost effective. Careful characterization of the detector performance will be undertaken. Issues such as detector stability, detector calibration, and count rate performance will be included. The R33 phase of the proposal will include building a dedicated small animal PET system using the cMiCE detector design. The preferred system design will have detector ring diameter of about 12 cm and an axial field of view approaching 10 cm. Monte Carlo simulation will be used to optimize the resolution-sensitivity (i.e., contrast versus noise) trade-off for estimation task performance. Additional simulations to quantify the effects of positron range for different radioisotopes will be conducted. The quantitative accuracy of images reconstructed with and without resolution recovery (e.g., positron range, photon acollinearity and detector response) will be studied. Finally, the feasibility of using the variation in the LRF to estimate depth of interaction for thick crystal designs will be investigated.