Cross-Stain Contrastive Learning for Paired Immunohistochemistry and Histopathology Slide Representation Learning

2025-12-03

AI Summary

To address inter-stain tissue misalignment in hematoxylin & eosin (H&E) and multiplex immunohistochemistry (IHC) whole-slide images (WSIs) caused by imperfect registration, this paper proposes a two-stage cross-stain contrastive learning (CSCL) framework. Methodologically, we construct the first five-stain paired WSI dataset, introduce a lightweight adapter and a cross-stain attention fusion module to achieve patch-level inter-stain feature alignment, and jointly optimize multiple instance learning (MIL) and global contrastive losses for self-supervised pretraining. This work presents the first framework enabling collaborative representation learning across H&E and multiplex IHC stains. Experiments demonstrate that the learned representations significantly improve performance on cancer subtype classification, IHC biomarker prediction, and survival analysis—achieving strong generalizability and cross-stain transferability.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Medical and Biological Imaging
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Universal, transferable whole-slide image (WSI) representations are central to computational pathology. Incorporating multiple markers (e.g., immunohistochemistry, IHC) alongside H&E enriches H&E-based features with diverse, biologically meaningful information. However, progress is limited by the scarcity of well-aligned multi-stain datasets. Inter-stain misalignment shifts corresponding tissue across slides, hindering consistent patch-level features and degrading slide-level embeddings. To address this, we curated a slide-level aligned, five-stain dataset (H&E, HER2, KI67, ER, PGR) to enable paired H&E-IHC learning and robust cross-stain representation. Leveraging this dataset, we propose Cross-Stain Contrastive Learning (CSCL), a two-stage pretraining framework with a lightweight adapter trained using patch-wise contrastive alignment to improve the compatibility of H&E features with corresponding IHC-derived contextual cues, and slide-level representation learning with Multiple Instance Learning (MIL), which uses a cross-stain attention fusion module to integrate stain-specific patch features and a cross-stain global alignment module to enforce consistency among slide-level embeddings across different stains. Experiments on cancer subtype classification, IHC biomarker status classification, and survival prediction show consistent gains, yielding high-quality, transferable H&E slide-level representations. The code and data are available at https://github.com/lily-zyz/CSCL.

Problem

• Addresses misalignment in multi-stain pathology datasets for consistent feature learning.
• Enhances H&E slide representations by integrating IHC-derived biological information.
• Improves computational pathology tasks like classification and survival prediction.

Innovation

• Two-stage pretraining with patch-wise contrastive alignment
• Cross-stain attention fusion for integrating patch features
• Slide-level alignment module for multi-stain consistency