Estimation of High Frame Rate Digital Subtraction Angiography Sequences at Low Radiation Dose

项目摘要 虽然数字减影血管造影术（DSA）提供了高分辨率的血流动态成像， 在动脉充盈和静脉引流期间通过大脑，其采集帧速率目前限于1-3 以最大限度地减少患者暴露于辐射。我们的长期目标是为实现 发展更容易解释的DSA成像技术。我们在这方面的总体目标 应用的目的是（i）开发一种能够产生高帧率DSA采集的新技术， 扰乱目前的临床实践，（ii）评估帧速率在缠结分类中的影响 供血和引流血管，以了解脑血管畸形。我们的核心假设是 来自实际低帧率DSA序列的图像可以被内插以生成新的中间图像 在任意高帧率下。该项目的理由是，这种技术可能会提供一个明确的 和可解释的成像工具，以临床医生，将有助于诊断和规划神经血管 诉讼中心假设将通过追求两个具体目标进行测试：1）开发一种技术， 从低帧率DSA序列生成高帧率DSA序列，以及2）测试高帧率DSA 在存在脑畸形的情况下缠绕的供血和引流血管的序列。根据第一项 为了达到这个目的，我们首先将序列分解为动脉、毛细血管和静脉相。接下来，我们将生成一个 扩散图来约束图像插值。最后，我们将使用非线性 基于对比度强度和预先建立的扩散图的插值方法。第二个目标，我们将 在缠绕的动脉和静脉上测试我们的方法，并使用各种相似性评估其准确性， 腐败度量我们认为，拟议的项目是创新的，因为它将有可能 从低帧率采集自动生成高帧率DSA序列，同时保持低帧率 辐射暴露我们的解决方案不会改变实际的临床常规，是一个软件解决方案， 硬件问题。拟议的项目意义重大，因为预计它将提供一个解决方案， 可以使用实际临床采集协议来生成高帧率DSA序列。结果 预计将产生重要的积极影响，因为它们最终将为发展中国家提供新的机会。 开发新的解释技术，以识别和治疗脑血管畸形。

Project Summary Although Digital Subtraction Angiography (DSA) provides high resolution, dynamic imaging of blood flow through the brain during arterial filling and venous drainage, its acquisition framerate is currently limited to 1-3 frames-per-seconds to minimize patients' exposure to radiation. Our long-term goal is to contribute toward the development of DSA imaging techniques that are more easily interpretable. Our overall objectives in this application are to (i) develop a new technology capable of producing high framerate DSA acquisition without disrupting current clinical practice, (ii) evaluate the influence of framerate in the classification of entangled feeding and draining vessels for the understanding of cerebrovascular malformations. Our central hypothesis is that images from actual low framerate DSA sequences can be interpolated to generate new in-between images at an arbitrary high framerate. The rationale for this project is that such technology will likely provide a clear and interpretable imaging tool to clinicians that will facilitate diagnosis and planning of neurovascular proceedings. The central hypothesis will be tested by pursuing two specific aims: 1) Develop a technique for generating high framerate DSA sequences from low framerate ones and 2) Test high framerate DSA sequences on entangled feeding and draining vessels in the presence of cerebral malformation. Under the first aim, we will first decompose the sequence into arterial, capillary and venous phases. Next, we will generate a diffusion map to constrain image interpolation. Finally, we will generate intermediate images using a non-linear interpolation method based on contrast intensity and the pre-built diffusion map. For the second aim, we will test our approach on entangled arteries and veins and evaluate its accuracy using various similarity and corruptibility metrics. The proposed project is innovative, in our opinion, because it will be possible to automatically generate high frame rate DSA sequences from low frame rate acquisitions while maintaining low radiation exposure. Our solution does not alter the actual clinical routine, and is a software solution to a hardware problem. The proposed project is significant, because it is expected to provide a solution by which a high framerate DSA sequence can be generated using actual clinical acquisition protocols. The results are expected to have an important positive impact because they will ultimately provide new opportunities for the development of novel interpretation techniques to identify and treat cerebrovascular malformations.