Sun Chang-yu, Chu Chun-yu, Liu Wan-yu, Hsu Edward W, Korenberg Julie R, Zhu Yue-min
CREATIS, CNRS UMR 5220, Inserm U1044, INSA of Lyon, University of Lyon, Villeurbanne 69100, France. CNRS LIA Metislab, Harbin Institute of Technology, Harbin 150001, People's Republic of China.
Phys Med Biol. 2015 Nov 7;60(21):8417-36. doi: 10.1088/0031-9155/60/21/8417. Epub 2015 Oct 14.
Diffusion tensor imaging and high angular resolution diffusion imaging are often used to analyze the fiber complexity of tissues. In these imaging techniques, the most commonly calculated metric is anisotropy, such as fractional anisotropy (FA), generalized anisotropy (GA), and generalized fractional anisotropy (GFA). The basic idea underlying these metrics is to compute the deviation from free or spherical diffusion. However, in many cases, the question is not really to know whether it concerns spherical diffusion. Instead, the main concern is to describe and quantify fiber complexity such as fiber crossing in a voxel. In this context, it would be more direct and effective to compute the deviation from a single fiber bundle instead of a sphere. We propose a new metric, called PEAM (PEAnut Metric), which is based on computing the deviation of orientation diffusion functions (ODFs) from a single fiber bundle ODF represented by a peanut. As an example, the proposed PEAM metric is used to classify intravoxel fiber configurations. The results on simulated data, physical phantom data and real brain data consistently showed that the proposed PEAM provides greater accuracy than FA, GA and GFA and enables parallel and complex fibers to be better distinguished.
扩散张量成像和高角分辨率扩散成像常用于分析组织的纤维复杂性。在这些成像技术中，最常计算的指标是各向异性，如分数各向异性（FA）、广义各向异性（GA）和广义分数各向异性（GFA）。这些指标背后的基本思想是计算与自由扩散或球形扩散的偏差。然而，在许多情况下，问题并非真的在于是否涉及球形扩散。相反，主要关注点是描述和量化纤维复杂性，例如体素中的纤维交叉。在这种情况下，计算与单个纤维束而非球体的偏差会更直接和有效。我们提出了一种新的指标，称为PEAM（花生指标），它基于计算方向扩散函数（ODF）与由花生表示的单个纤维束ODF的偏差。例如，所提出的PEAM指标用于对体素内纤维构型进行分类。在模拟数据、物理体模数据和真实脑数据上的结果一致表明，所提出的PEAM比FA、GA和GFA具有更高的准确性，并且能够更好地区分平行和复杂纤维。
