Raniga Parnesh, Bourgeat Pierrick, Villemagne Victor, O'Keefe Graeme, Rowe Christopher, Ourselin Sébastien
BioMedIA Lab, e-Health Research Centre, CSIRO ICT Centre, Brisbane, Australia.
Med Image Comput Comput Assist Interv. 2007;10(Pt 1):228-35. doi: 10.1007/978-3-540-75757-3_28.
With the advent of biomarkers such as 11C-PIB and the increase in use of PET, automated methods are required for processing and analyzing datasets from research studies and in clinical settings. A common preprocessing step is the calculation of standardized uptake value ratio (SUVR) for inter-subject normalization. This requires segmented grey matter (GM) for VOI refinement. However 11C-PIB uptake is proportional to amyloid build up leading to inhomogeneities in intensities, especially within GM. Inhomogeneities present a challenge for clustering and pattern classification based approaches to PET segmentation as proposed in current literature. In this paper we modify a MR image segmentation technique based on expectation maximization for 11C-PIB PET segmentation. A priori probability maps of the tissue types are used to initialize and enforce anatomical constraints. We developed a Bézier spline based inhomogeneity correction techniques that is embedded in the segmentation algorithm and minimizes inhomogeneity resulting in better segmentations of 11C-PIB PET images. We compare our inhomogeneity with a global polynomial correction technique and validate our approach using co-registered MRI segmentations.
随着诸如11C-PIB等生物标志物的出现以及PET使用的增加，研究和临床环境中处理和分析数据集需要自动化方法。一个常见的预处理步骤是计算标准化摄取值比率（SUVR）以进行受试者间归一化。这需要分割灰质（GM）以细化感兴趣区域（VOI）。然而，11C-PIB摄取与淀粉样蛋白积累成比例，导致强度不均匀，尤其是在GM内。不均匀性给当前文献中提出的基于聚类和模式分类的PET分割方法带来了挑战。在本文中，我们修改了基于期望最大化的MR图像分割技术用于11C-PIB PET分割。组织类型的先验概率图用于初始化和强化解剖学约束。我们开发了一种基于贝塞尔样条的不均匀性校正技术，该技术嵌入到分割算法中，并最小化不均匀性，从而实现对11C-PIB PET图像更好的分割。我们将我们的不均匀性校正方法与全局多项式校正技术进行比较，并使用配准的MRI分割来验证我们的方法。
