Visualizing Propagator-Based Diffusion Imaging Data

扩散磁共振成像及其许多变体在研究和临床中都被用于对人脑白质进行成像。近年来，随着技术的进步，不仅可以用高角度分辨率测量扩散，而且还可以用不同强度的扩散加权测量扩散。这使得估计包含有关组织微结构的附加信息的三维扩散传播因子成为可能。然而，将这种错综复杂的数据可视化的方法很少。我们的项目首先将通过实现适合于高效数据存储、内插和特征提取的表示来为交互式扩散传播子可视化奠定基础。在此基础上，我们将在不断增加的空间尺度上设计和实施可视化技术：从局部(字形)到单个解剖结构(轨迹图)和全局结构(直接体绘制)。这些方法的灵感来自于已成功用于可视化扩散张量场的技术。然而，扩散传播子额外的径向和高角度分辨率将需要实质性的扩展。我们将特别注意考虑我们的合奏和比较可视化技术的不确定性和适用性。最后，我们将与临床合作伙伴密切合作，在真实数据上测试我们的技术。

Diffusion Magnetic Resonance Imaging and its many variants are firmly established for imaging the white matter of the human brain, both in research and in the clinic. In recent years, technical progress has increasingly made it feasible to measure diffusion not just with high angular resolution, but in addition with varying strengths of diffusion weighting. This makes it possible to estimate a three-dimensional diffusion propagator, which contains additional information about tissue microstructure. However, very few approaches to visualizing this intricate data exist. Our project will first lay the foundation for interactive diffusion propagator visualization by implementing representations that are suitable for efficient data storage, interpolation, and feature extraction. Building on this, we will design and implement visualization techniques at increasing spatial scales: From local (glyphs) to individual anatomical structures (tractography) and global structures (direct volume rendering). These methods are inspired by techniques that have been successfully used for visualizing diffusion tensor fields. However, the additional radial and the high angular resolution of diffusion propagators will require substantial extensions. We will pay special attention to account for uncertainty and suitability of our techniques for ensemble and comparative visualization. Finally, we will test our techniques on real-world data in close collaboration with a clinical partner.