MiCo: Multiple Instance Learning with Context-Aware Clustering for Whole Slide Image Analysis

2025-06-22

AI Summary

Whole-slide images (WSIs) exhibit pronounced spatial heterogeneity, causing morphologically similar tissue regions to be dispersed across anatomically distant locations—posing a fundamental challenge for conventional multi-instance learning (MIL) frameworks to capture long-range inter-region dependencies. To address this, we propose a context-aware clustering-based MIL framework. It introduces a clustering-path module that dynamically links spatially distant yet histologically homogeneous instances, and a clustering-reduction module that enhances semantic interaction among clusters to mitigate semantic fragmentation. Further, we integrate semantic-anchor-guided feature optimization, dynamic similarity matching, and redundant-cluster merging. Evaluated on nine large-scale cancer datasets across diagnostic and prognostic tasks, our method consistently outperforms state-of-the-art approaches, demonstrating superior capability in modeling spatial heterogeneity and strong generalizability across diverse cancer types and clinical endpoints.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

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

Abstract

Multiple instance learning (MIL) has shown significant promise in histopathology whole slide image (WSI) analysis for cancer diagnosis and prognosis. However, the inherent spatial heterogeneity of WSIs presents critical challenges, as morphologically similar tissue types are often dispersed across distant anatomical regions. Conventional MIL methods struggle to model these scattered tissue distributions and capture cross-regional spatial interactions effectively. To address these limitations, we propose a novel Multiple instance learning framework with Context-Aware Clustering (MiCo), designed to enhance cross-regional intra-tissue correlations and strengthen inter-tissue semantic associations in WSIs. MiCo begins by clustering instances to distill discriminative morphological patterns, with cluster centroids serving as semantic anchors. To enhance cross-regional intra-tissue correlations, MiCo employs a Cluster Route module, which dynamically links instances of the same tissue type across distant regions via feature similarity. These semantic anchors act as contextual hubs, propagating semantic relationships to refine instance-level representations. To eliminate semantic fragmentation and strengthen inter-tissue semantic associations, MiCo integrates a Cluster Reducer module, which consolidates redundant anchors while enhancing information exchange between distinct semantic groups. Extensive experiments on two challenging tasks across nine large-scale public cancer datasets demonstrate the effectiveness of MiCo, showcasing its superiority over state-of-the-art methods. The code is available at https://github.com/junjianli106/MiCo.

Problem

• Address spatial heterogeneity in whole slide images
• Enhance cross-regional intra-tissue correlations
• Strengthen inter-tissue semantic associations

Innovation

• Context-aware clustering for WSI analysis
• Cluster Route module links distant tissue instances
• Cluster Reducer eliminates redundant semantic anchors