Alzheimer's disease (AD) is a neurodegenerative disorder, considered a disconnection syndrome with regional molecular pattern abnormalities quantifiable by the aid of PET imaging. Solutions for accurate quantification of network dysfunction are scarce. We evaluate the extent to which PET molecular markers reflect quantifiable network metrics derived through the graph theory framework and how partial volume effects (PVE)-correction (PVEc) affects these PET-derived metrics 75 AD patients and 126 cognitively normal older subjects (CN). Therefore our goal is twofold: 1) to evaluate the differential patterns of [18F]FDG- and [18F]AV45-PET data to depict AD pathology; and ii) to analyse the effects of PVEc on global uptake measures of [18F]FDG- and [18F]AV45-PET data and their derived covariance network reconstructions for differentiating between patients and normal older subjects. Network organization patterns were assessed using graph theory in terms of "degree", "modularity", and "efficiency". PVEc evidenced effects on global uptake measures that are specific to either [18F]FDG- or [18F]AV45-PET, leading to increased statistical differences between the groups. PVEc was further shown to influence the topological characterization of PET-derived covariance brain networks, leading to an optimised characterization of network efficiency and modularisation. Partial-volume effects correction improves the interpretability of PET data in AD and leads to optimised characterization of network properties for organisation or disconnection.

中文翻译：

---

阿尔茨海默氏病的代谢和淀粉样蛋白PET网络重组：差异模式和部分体积效应。

阿尔茨海默氏病（AD）是一种神经退行性疾病，被认为是一种断路综合征，其局部分子模式异常可通过PET成像进行量化。缺乏精确的网络功能障碍定量解决方案。我们评估了PET分子标记在多大程度上反映了通过图论框架得出的可量化网络指标，以及部分体积效应（PVE）-校正（PVEc）如何影响这些PET指标，包括75位AD患者和126名认知正常的老年受试者（CN） 。因此，我们的目标是双重的：1）评估[18F] FDG-和[18F] AV45-PET数据的差异模式以描述AD病理；ii）分析PVEc对[18F] FDG-和[18F] AV45-PET数据的整体摄取量度及其派生的协方差网络重构的影响，以区分患者和正常老年受试者。使用图论根据“程度”，“模块化”和“效率”评估网络组织模式。PVEc证明了对[18F] FDG-或[18F] AV45-PET特有的总体摄取指标的影响，导致两组之间的统计学差异增加。PVEc被进一步证明会影响PET衍生的协方差大脑网络的拓扑特征，从而导致网络效率和模块化的优化特征。