Representation learning of pathology whole-slide images (WSIs) has been has primarily relied on weak supervision with Multiple Instance Learning (MIL). However, the slide representations resulting from this approach are highly tailored to specific clinical tasks, which limits their expressivity and generalization, particularly in scenarios with limited data. Instead, we hypothesize that morphological redundancy in tissue can be leveraged to build a task-agnostic slide representation in an unsupervised fashion. To this end, we introduce PANTHER, a prototype-based approach rooted in the Gaussian mixture model that summarizes the set of WSI patches into a much smaller set of morphological prototypes. Specifically, each patch is assumed to have been generated from a mixture distribution, where each mixture component represents a morphological exemplar. Utilizing the estimated mixture parameters, we then construct a compact slide representation that can be readily used for a wide range of downstream tasks. By performing an extensive evaluation of PANTHER on subtyping and survival tasks using 13 datasets, we show that 1) PANTHER outperforms or is on par with supervised MIL baselines and 2) the analysis of morphological prototypes brings new qualitative and quantitative insights into model interpretability.

翻译：全切片病理图像（WSI）的表征学习主要依赖基于多实例学习（MIL）的弱监督方法。然而，该方法生成的切片表征高度适配特定临床任务，导致其表达能力和泛化性受限，尤其在数据稀缺场景中表现尤为突出。为此，我们提出利用组织形态学冗余性，以无监督方式构建任务无关的切片表征。基于这一假设，我们设计了基于高斯混合模型的原型学习框架PANTHER，该框架将WSI的斑块集合归纳为更精简的形态学原型簇。具体而言，每个斑块被视为从混合分布中生成，其中每个混合分量对应一个形态学原型。利用估计的混合参数，我们构建了可直接用于各类下游任务的紧凑型切片表征。通过在13个数据集上对分型与生存预测任务进行系统评估，我们证明：1）PANTHER的性能优于或持平于有监督MIL基线方法；2）形态学原型分析为模型可解释性提供了全新的定性与定量洞见。