Simulation Tools for 3D and 4D CT and Dosimetry

点击翻译按钮获取中文摘要

Abstract Photon-counting CT (PCCT) is a major technological advance in CT imaging. Using photon-counting instead of current energy-integrating detectors, PCCT can offer superior performance in terms of spatial resolution, artifact reduction, and most notably, material decomposition. PCCT’s energy differentiation utility offers an ability to more precisely distinguish different materials and optimize and expand the use of contrast agents in CT. With these abilities, PCCT can significantly facilitate quantitative imaging, reduce radiation exposure, and enable revolutionary new applications in functional and physiological imaging beyond existing CT techniques. To realize the full potential of PCCT in clinical practice, the technology needs comprehensive assessments and application-based optimizations. Effective design and deployment of PCCT depends on many design and use choices that should be made in view of the eventual clinical utility. Making these choices requires large scale trials on actual patients. However, such trials are challenging, considering the need to make many decisions prior to prototyping, the limited numbers of prototype PCCT scanners available today, and the often-unknown ground-truth in the patient images. Even for existing prototype systems, many decisions require repetitive trials with multiple acquisitions. This is both unethical and impractical considering radiation safety concerns and costs. These challenges can be overcome by utilizing virtual imaging trials (VITs) using computerized patients and imaging models. VITs provide an efficient means with which to determine the most effective and optimized design and use of imaging technologies with complete control over the study design. In our prior funded project, we developed a VIT framework to evaluate standard energy-integrating detector CT technologies. In this project, we expand the applicability of this framework to photon-counting detector CT. Specifically, we enhance our computational XCAT phantoms to model the necessary higher-resolution detail including normal and abnormal tissue heterogeneities and intra-organ contrast perfusion diversity across populations (Aim 1). To image the phantoms, we develop the first PCCT simulator capable of mimicking existing and emerging prototypes (Aim 2). The enhanced VIT framework will provide the essential foundation with which to comprehensively evaluate and optimize PCCT technologies and applications. In Aim 3, we assess and optimize the use of PCCT for morphological, textural, and compositional quantification in select oncologic and cardiac applications, two leading health detriments in the US where PCCT can offer a notable impact. The results will be the first of their kind in comprehensively evaluating the task-based merits and capabilities of PCCT, determining optimum dose per patient size for PCCT imaging of patients for cancerous lesions and cardiac plaque/stenoses, and helping to establish the effective utility of PCCT in clinical care.