PC-MIL: Decoupling Feature Resolution from Supervision Scale in Whole-Slide Learning

2026-04-13

AI Summary

This study addresses the mismatch between supervision granularity and clinical diagnostic scale in traditional multiple instance learning (MIL), which relies solely on slide-level labels and fails to model anatomically meaningful structures. To resolve this, the authors propose Progressive-Context MIL (PC-MIL), a novel framework that decouples supervision scale from feature resolution for the first time. Operating on fixed 20× features, PC-MIL introduces millimeter-scale regional supervision and progressively integrates it with slide-level labels to explicitly model anatomical context. Notably, this approach enhances clinical interpretability and generalization without requiring pixel-level annotations or changes in magnification. Experiments on 1,476 prostate whole-slide images demonstrate that modest regional supervision substantially improves cross-context performance, maintaining global accuracy while stabilizing both regional and slide-level evaluation metrics.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Multiagent Systems: Multiagent Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Whole-slide image (WSI) classification in computational pathology is commonly formulated as slide-level Multiple Instance Learning (MIL) with a single global bag representation. However, slide-level MIL is fundamentally underconstrained: optimizing only global labels encourages models to aggregate features without learning anatomically meaningful localization. This creates a mismatch between the scale of supervision and the scale of clinical reasoning. Clinicians assess tumor burden, focal lesions, and architectural patterns within millimeter-scale regions, whereas standard MIL is trained only to predict whether "somewhere in the slide there is cancer." As a result, the model's inductive bias effectively erases anatomical structure. We propose Progressive-Context MIL (PC-MIL), a framework that treats the spatial extent of supervision as a first-class design dimension. Rather than altering magnification, patch size, or introducing pixel-level segmentation, we decouple feature resolution from supervision scale. Using fixed 20x features, we vary MIL bag extent in millimeter units and anchor supervision at a clinically motivated 2mm scale to preserve comparable tumor burden and avoid confounding scale with lesion density. PC-MIL progressively mixes slide- and region-level supervision in controlled proportions, enabling explicit train-context x test-context analysis. On 1,476 prostate WSIs from five public datasets for binary cancer detection, we show that anatomical context is an independent axis of generalization in MIL, orthogonal to feature resolution: modest regional supervision improves cross-context performance, and balanced multi-context training stabilizes accuracy across slide and regional evaluation without sacrificing global performance. These results demonstrate that supervision extent shapes MIL inductive bias and support anatomically grounded WSI generalization.

Problem

• Multiple Instance Learning
• Whole-slide image
• Supervision scale
• Anatomical localization
• Computational pathology

Innovation

• Multiple Instance Learning
• supervision scale
• whole-slide image
• anatomical context
• computational pathology