Morphological Prototyping for Unsupervised Slide Representation Learning in Computational Pathology

• Main Authors: Song, Andrew H., Chen, Richard J., Ding, Tong, Williamson, Drew F.K., Jaume, Guillaume, Mahmood, Faisal
• Format: Conference Proceeding
• Language: English
• Subjects: Analytical models; Codes; Computational modeling; computational pathology; Computer vision; Pathology; prototype; Redundancy; Representation learning; set representation learning
• ISSN: 2575-7075
• DOI: 10.1109/CVPR52733.2024.01099

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 distri-bution, 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 super-vised MIL baselines and 2) the analysis of morphological prototypes brings new qualitative and quantitative in-sights into model interpretability. The code is available at https://github.com/mahmoodlab/Panther.