In digital pathology, the spatial context of cells is important for cell classification, cancer diagnosis and prognosis. To model such complex cell context, however, is challenging. Cells form different mixtures, lineages, clusters and holes. To model such structural patterns in a learnable fashion, we introduce several mathematical tools from spatial statistics and topological data analysis. We incorporate such structural descriptors into a deep generative model as both conditional inputs and a differentiable loss. This way, we are able to generate high quality multi-class cell layouts for the first time. We show that the topology-rich cell layouts can be used for data augmentation and improve the performance of downstream tasks such as cell classification.
翻译:数字病理学中,细胞的空间上下文对于细胞分类、癌症诊断和预后是十分重要的。然而,如何建模这种复杂细胞结构是具有挑战性的。细胞形成不同的混合物、宗系、簇和空洞。为了以可学习的方式建模这种结构模式,我们引入了空间统计学和拓扑数据分析的一些数学工具。我们将这些结构描述符合并到了深度生成模型中,作为条件输入和可微分损失。通过这种方式,我们能够首次生成高质量的多类细胞布局。我们展示了富有拓扑学的细胞布局可用于数据增强,并提高下游任务(例如细胞分类)的性能。