Quantitative ECT for Complex Imaging Tasks

描述(由申请人提供)：拟议研究的长期目标是改进从具有挑战性的放射性核素的发射断层扫描图像中获取、重建和提取定量信息，并使用任务相关标准来评估这些改进。在本项目期间，我们集中力量开展了111In和67Cu等复杂放射性核素，以及111In和99mTc的同步双放射性核素成像。在下一个项目期间，我们将完善我们对这些放射性核素的研究；然而，我们还将朝着三个新方向扩展工作，以解决以下方面的重要技术挑战：小鼠临床前发射断层成像，用于放射性核素治疗的贝塔发射体的韧致辐射成像，以及减少由于物理因素造成的系统性误差，这些因素目前限制了靶向放射性核素治疗计划的准确性和精确度。我们将考虑的成像任务包括估计肿瘤或感染区域的活性浓度。我们将使用模拟数据以及幻影、动物和/或患者数据来衡量我们的新方法在任务绩效方面的改善，并将所取得的绩效与理论界限进行比较。我们将考虑与表达生长抑素受体的肿瘤的放射治疗计划相关的临床任务，以及90Y标记微球治疗肝肿瘤和转移瘤的临床任务，以及骨髓炎的诊断。我们将把我们的SPECT准直器设计工作扩展到同时进行双核素成像的情况，并解决基于投影数据集的准直器优化是否足够，或者准直器是否应该与重建算法联合优化的悬而未决的问题。我们还将开发和评估一种用于韧致辐射SPECT的重建技术，以及一种基于模型的程序，用于校正SPECT-CT和PET-CT图像的“部分体积效应”。最后，我们将比较基于蒙特卡罗的迭代重建算法和使用双示踪数据的定量平面成像方法诊断骨髓炎的性能。与公共卫生相关：拟议的研究将解决使几种放射性核素的发射断层成像复杂化的技术挑战，这些放射性核素对癌症的诊断、治疗计划和治疗以及骨感染的诊断都有价值。这项研究将带来的洞察力和改进的成像技术将导致对这些疾病的更好诊断和管理。

DESCRIPTION (provided by applicant): The long-term goals of the proposed research are to improve the acquisition, reconstruction, and extraction of quantitative information from emission tomographic images of challenging radionuclides and to assess these improvements using task-dependent criteria. During the current project period, we have concentrated our efforts on complex radionuclides, e.g., 111In and 67Cu, as well as on simultaneous dual- radionuclide imaging of 111In and 99mTc. During the next project period, we will refine our research with these radionuclides; however, we will also extend the work in three new directions in order to address important technical challenges in pre-clinical emission tomographic imaging of mice, bremsstrahlung imaging of beta emitters used for radionuclide therapy, and reduction of systematic errors due to physical factors that currently limit the accuracy and precision of treatment planning in targeted radionuclide therapy. The imaging tasks that we will consider involve estimation of activity concentration in tumors or in regions of infection. We will measure, using simulated data as well as phantom, animal, and/or patient data, the improvements in task performance resulting from our new methodologies, and compare the performance achieved to theoretical bounds. We will consider clinical tasks related to treatment planning for radiotherapy of tumors expressing somatostatin receptors, as well as to treatment of hepatic tumors and metastases with 90Y- labeled microspheres, and to the diagnosis of osteomyelitis. We will extend our SPECT collimator design work to the case of simultaneous dual-radionuclide imaging, and to address the unresolved issue of whether collimator optimization on the basis of projection datasets is sufficient, or whether the collimator should be jointly optimized with the reconstruction algorithm. We will also develop and assess a reconstruction technique for bremsstrahlung SPECT, as well as a model- based procedure for correcting SPECT-CT and PET-CT images for the "partial volume effect." Finally, we will compare the performance of a Monte-Carlo based iterative reconstruction algorithm to that of a quantitative planar imaging approach for the diagnosis of osteomyelitis using dual-tracer data. PUBLIC HEALTH RELEVANCE: The proposed research will address technical challenges which complicate emission tomographic imaging of several radionuclides which are of value for diagnosis, treatment planning, and therapy of cancer, as well as for diagnosis of bone infection. The insights and improved imaging techniques which will emerge from this research will lead to better diagnosis and management of these diseases.