K-Stain: Keypoint-Driven Correspondence for H&E-to-IHC Virtual Staining

2025-11-10

AI Summary

To address spatial misalignment challenges in H&E-to-IHC virtual staining caused by tissue section displacement, this paper proposes K-Stain, the first hierarchical keypoint-driven cross-modal image translation framework. Methodologically, it introduces three synergistic modules: (i) a Hierarchical Spatial Keypoint Detector (HSKD) for robust anatomical keypoint localization; (ii) a Keypoint-aware Enhancement Generator (KEG) that fuses keypoint-guided spatial-semantic features; and (iii) a Keypoint-Guided Discriminator (KGD) enforcing structural consistency. By explicitly modeling anatomical correspondences and incorporating contextual information from adjacent sections, K-Stain significantly improves spatial alignment accuracy and fine-grained tissue structure preservation. Extensive experiments on multiple benchmarks demonstrate that K-Stain outperforms state-of-the-art methods in PSNR, SSIM, and expert pathologist evaluations, validating the effectiveness and generalizability of keypoint guidance in computational pathology virtual staining.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Kernel Methods
• Knowledge Representation and Reasoning: Knowledge Engineering

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Virtual staining offers a promising method for converting Hematoxylin and Eosin (H&E) images into Immunohistochemical (IHC) images, eliminating the need for costly chemical processes. However, existing methods often struggle to utilize spatial information effectively due to misalignment in tissue slices. To overcome this challenge, we leverage keypoints as robust indicators of spatial correspondence, enabling more precise alignment and integration of structural details in synthesized IHC images. We introduce K-Stain, a novel framework that employs keypoint-based spatial and semantic relationships to enhance synthesized IHC image fidelity. K-Stain comprises three main components: (1) a Hierarchical Spatial Keypoint Detector (HSKD) for identifying keypoints in stain images, (2) a Keypoint-aware Enhancement Generator (KEG) that integrates these keypoints during image generation, and (3) a Keypoint Guided Discriminator (KGD) that improves the discriminator's sensitivity to spatial details. Our approach leverages contextual information from adjacent slices, resulting in more accurate and visually consistent IHC images. Extensive experiments show that K-Stain outperforms state-of-the-art methods in quantitative metrics and visual quality.

Problem

• Converting H&E images to IHC images virtually
• Overcoming misalignment in tissue slice spatial information
• Enhancing IHC image fidelity using keypoint-driven correspondence

Innovation

• Uses keypoints for spatial correspondence alignment
• Integrates keypoint-aware enhancement in image generation
• Employs hierarchical detector and guided discriminator components