Wavelet-aided parametric mapping of cerebral dopamine D2 receptors using the high affinity PET radioligand [11C]FLB 457
Wavelet-aided parametric mapping of cerebral dopamine D2 receptors using the high affinity PET radioligand [11C]FLB 457
复制标题
DOI：
10.1006/nimg.2002.1152
复制
发表时间：
2002-09-01
期刊：
影响因子：
5.7
通讯作者：
Gulyás, B
中科院分区：
文献类型：
--
作者：
Cselényi, Z;Olsson, H;Gulyás, B
通过正电子发射断层扫描（PET）和合适的放射性配体研究人类神经受体系统已被证明在正常脑功能和神经精神疾病的病理生理学和治疗的研究中具有重要意义。一个长期确定的目标是产生详细的参数图，显示整个人脑在体内的神经受体结合参数。小波滤波器的应用最近被提出作为一种解决方案，以处理固有的低信噪比的PET图像。在本研究中，我们应用小波方法从10名健康受试者谁与[C-11]FLB 457检查获得的数据。这种高亲和力的多巴胺D-2受体拮抗剂提供了一个信号，从一系列的区域与百倍的受体密度的差异，因此应该是适合的小波方法的评价。为了交叉验证的目的，数据进行了分析与四种方法：传统的区域的利益（ROI）为基础的分析，基于像素的分析和两个变量的小波辅助分析。在这两种变体中，小波滤波器在空间上被应用，但是在一种情况下使用二维滤波器，而在另一种情况下使用三维滤波器。所有方法均使用相同的线性图形结合势（BP）估计步骤，并通过计算代表性ROI的平均BP，将三种参数映射技术的结果与基于参考ROI的方法进行比较。基于像素和二维小波的方法产生了高度相关，但系统较低的值相比，参考ROI为基础的方法。利用三维小波滤波器的方法产生的BP地图与区域平均值密切匹配的ROI为基础的方法的值。结果表明，三维空间小波滤波与建立的参数估计程序相结合，提供了详细的，准确的地图的放射性配体结合参数。这种地图可用于在次区域一级对结合参数进行个体间或多条件比较。(C)2002 Elsevier Science（美国）。
The study of human neuroreceptor systems by means of positron emission tomography (PET) and suitable radioligands has proven to be of great importance in research on normal brain functions and the pathophysiology and treatment of neuropsychiatric disorders. A for long identified goal is to produce detailed parametric maps of showing neuroreceptor binding parameters for the entire human brain in vivo. The application of wavelet filters has recently been proposed as a solution to handle the inherently low signal-to-noise ratio of PET images. In the present study we applied the wavelet approach to data obtained from 10 healthy subjects who were examined with [C-11]FLB 457. This high affinity dopamine D-2-receptor antagonist provides a signal from a range of regions with a hundredfold difference in receptor density and should thus be suitable for evaluation of the wavelet approach. For cross-validation purposes the data were analysed with four methods: a traditional region-of-interest (ROI) based analysis, a pixel-based analysis and two variants of wavelet-aided analyses. In both variants the wavelet filter was spatially applied, but a two-dimensional filter was used in one case and a three-dimensional one in the other. The same linear-graphical binding potential (BP) estimation step was used for all methods and the results of the three parametric mapping techniques were compared to the reference ROI-based method by calculating the average BP of representative ROIs. The pixel-based and the two-dimensional-wavelet-based methods yielded highly correlated but systematically lower values when compared to the reference ROI-based method. The approach utilising three-dimensional wavelet filters yielded BP maps with regional averages closely matching the values of the ROI-based method. The results show that the combination of three-dimensional spatial wavelet filtering with established parameter estimation procedures provides detailed, accurate maps of radioligand binding parameters. Such maps can be used for in inter-individual or multi-condition comparisons of binding parameters at subregional levels. (C) 2002 Elsevier Science (USA).