A Generative Shape Compositional Framework: Towards Representative Populations of Virtual Heart Chimaeras-猫眼课题宝

DOI：

10.48550/arxiv.2210.01607

发表时间：

2022

期刊：

ArXiv

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Haoran Dou;S. Virtanen;N. Ravikumar;Alejandro F Frangi

—生成能够捕获足够变异性同时保持合理性的虚拟解剖群体对于进行医疗设备的计算机模拟试验至关重要。然而，并非所有感兴趣的解剖形状都始终适用于人群中的每个个体。所采用和可用的影像检查和方式可能因人而异。不同的成像方式可能具有不同的视野，对来自不同组织/器官的信号敏感，或两者兼而有之。因此，群体中的个体通常可以获得缺失/部分重叠的解剖信息。我们引入了一种用于复杂解剖结构的生成形状模型，可以从不成对的数据集的数据集中学习，即复杂中的每个子结构都来自于同一群体的不相交受试者的缺失或部分重叠子结构的数据集。所提出的生成模型可以合成完整的整体复杂形状组件，创造出虚拟嵌合体，而不是自然的人类嵌合体。我们应用这个框架从全心形状组件的数据库构建虚拟嵌合体，每个嵌合体都为心脏子结构提供样本。具体来说，我们提出了一种基于图神经网络的生成形状组合框架，该框架包括两个组件：一个部分感知的生成形状模型，它捕获训练群体中每个感兴趣结构观察到的形状变化；以及一个空间合成网络，它将前者合成的结构组装/合成为多部分形状组件（即虚拟嵌合体）。我们还提出了一种新颖的自监督学习方案，使空间合成网络能够使用部分重叠的数据和弱标签进行训练。我们使用英国生物库中提供的心脏磁共振图像衍生的心脏结构形状来训练和验证我们的方法。当使用完整和部分重叠的数据进行训练时，我们的方法在通用性和特异性方面明显优于基于 PCA 的形状模型（使用完整数据进行训练）。这证明了所提出的方法的优越性

—Generating virtual populations of anatomy that capture sufﬁcient variability while remaining plausible is essential for conducting in-silico trials of medical devices. However, not all anatomical shapes of interest are always available for each individual in a population. Imaging examinations and modalities employed and available can vary across individuals. Different imaging modalities may have different ﬁelds of view, be sensitive to signals from different tissues/organs, or both. Hence, missing/partially-overlapping anatomical information is often available across individuals in a population. We introduce a generative shape model for complex anatomical structures, learnable from datasets of unpaired datasets, i.e. where each substructure in the complex comes from datasets that have missing or partially overlapping substructures from disjoint subjects of the same population. The proposed generative model can synthesise complete whole-complex shape assemblies coined virtual chimaeras , as opposed to natural human chimaeras. We applied this framework to build virtual chimaeras from databases of whole-heart shape assemblies that each contribute samples for heart substructures. Speciﬁcally, we propose a graph neural network-based generative shape compositional framework which comprises two components - a part-aware generative shape model which captures the variability in shape observed for each structure of interest in the training population; and a spatial composition network which assembles/composes the structures synthesised by the former into multi-part shape assemblies (viz. virtual chimaeras ). We also propose a novel self-supervised learning scheme that enables the spatial composition network to be trained with partially overlapping data and weak labels. We trained and validated our approach using shapes of cardiac structures derived from cardiac magnetic resonance images available in the UK Biobank. When trained with complete and partially overlapping data, our approach signiﬁcantly outperforms a PCA-based shape model (trained with complete data) in terms of generalisability and speciﬁcity. This demonstrates the superiority of the proposed approach as

A Generative Shape Compositional Framework: Towards Representative Populations of Virtual Heart Chimaeras

文献摘要

相似文献