Prototype Instance-semantic Disentanglement with Low-rank Regularized Subspace Clustering for WSIs Explainable Recognition

TL;DR

This paper introduces PID-LRSC, an innovative framework that disentangles instance semantics in WSIs using low-rank regularized subspace clustering and contrastive learning, improving interpretability and diagnostic reliability.

Contribution

The paper proposes a novel end-to-end framework combining low-rank subspace clustering and contrastive learning for semantic disentanglement in WSIs, addressing instance and semantic entanglement issues.

Findings

Outperforms state-of-the-art methods on multicentre pathology datasets.

Provides clearer instance semantics for better interpretability.

Enhances diagnostic reliability in pathological analysis.

Abstract

The tumor region plays a key role in pathological diagnosis. Tumor tissues are highly similar to precancerous lesions and non tumor instances often greatly exceed tumor instances in whole slide images (WSIs). These issues cause instance-semantic entanglement in multi-instance learning frameworks, degrading both model representation capability and interpretability. To address this, we propose an end-to-end prototype instance semantic disentanglement framework with low-rank regularized subspace clustering, PID-LRSC, in two aspects. First, we use secondary instance subspace learning to construct low-rank regularized subspace clustering (LRSC), addressing instance entanglement caused by an excessive proportion of non tumor instances. Second, we employ enhanced contrastive learning to design prototype instance semantic disentanglement (PID), resolving semantic entanglement caused by the high similarity between tumor and precancerous tissues. We conduct extensive experiments on multicentre pathology datasets, implying that PID-LRSC outperforms other SOTA methods. Overall, PID-LRSC provides clearer instance semantics during decision-making and significantly enhances the reliability of auxiliary diagnostic outcomes.