MacNet: An End-to-End Manifold-Constrained Adaptive Clustering Network for Interpretable Whole Slide Image Classification

2026-02-16

AI Summary

This work addresses the limitations of existing whole-slide image classification methods in interpretability and semantic clarity of learned features, particularly the susceptibility of clustering to high-dimensional noise and semantically ambiguous cluster centers. To overcome these challenges, the authors propose an end-to-end multi-instance learning framework that uniquely integrates Grassmann manifold re-embedding with manifold-adaptive clustering, complemented by a prior knowledge–guided proxy instance annotation mechanism to emphasize pathologically relevant regions. By leveraging the geometric structure of the Grassmann manifold, the approach enhances clustering robustness and achieves significant improvements in both grading accuracy and model interpretability across multi-center datasets, while enabling efficient and principled end-to-end feature learning.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Interpretability, Explainability, and Transparency
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Whole slide images (WSIs) are the gold standard for pathological diagnosis and sub-typing. Current main-stream two-step frameworks employ offline feature encoders trained without domain-specific knowledge. Among them, attention-based multiple instance learning (MIL) methods are outcome-oriented and offer limited interpretability. Clustering-based approaches can provide explainable decision-making process but suffer from high dimension features and semantically ambiguous centroids. To this end, we propose an end-to-end MIL framework that integrates Grassmann re-embedding and manifold adaptive clustering, where the manifold geometric structure facilitates robust clustering results. Furthermore, we design a prior knowledge guiding proxy instance labeling and aggregation strategy to approximate patch labels and focus on pathologically relevant tumor regions. Experiments on multicentre WSI datasets demonstrate that: 1) our cluster-incorporated model achieves superior performance in both grading accuracy and interpretability; 2) end-to-end learning refines better feature representations and it requires acceptable computation resources.

Problem

• whole slide image classification
• interpretable AI
• clustering ambiguity
• high-dimensional features
• multiple instance learning

Innovation

• manifold-constrained clustering
• Grassmann re-embedding
• interpretable MIL
• proxy instance labeling
• end-to-end WSI classification