Adaptive H&E-IHC information fusion staining framework based on feature extra

2025-02-27

AI Summary

Existing H&E-to-IHC staining translation methods suffer from two key limitations: (1) reliance on weak pixel-level features, leading to pathological structural information loss; and (2) absence of pixel-accurate paired data, rendering standard L1 loss ineffective. To address these, we propose WaveColor—a novel framework featuring (i) a multi-scale wavelet-based feature extraction module (VMFE) that jointly encodes discriminative spatial and frequency-domain representations; and (ii) a high-dimensional contrastive learning–driven dual-encoder architecture that enables unsupervised, adaptive alignment of H&E and IHC feature spaces without precise pixel-wise supervision, augmented by a feature-enhancement weighted loss. Evaluated on multiple public histopathological datasets, WaveColor significantly improves staining fidelity and tissue structural consistency (PSNR ↑3.2 dB, SSIM ↑0.08). The source code is publicly available.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Humans and AI: Human-in-the-loop Machine Learning

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
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Immunohistochemistry (IHC) staining plays a significant role in the evaluation of diseases such as breast cancer. The H&E-to-IHC transformation based on generative models provides a simple and cost-effective method for obtaining IHC images. Although previous models can perform digital coloring well, they still suffer from (i) coloring only through the pixel features that are not prominent in HE, which is easy to cause information loss in the coloring process; (ii) The lack of pixel-perfect H&E-IHC groundtruth pairs poses a challenge to the classical L1 loss.To address the above challenges, we propose an adaptive information enhanced coloring framework based on feature extractors. We first propose the VMFE module to effectively extract the color information features using multi-scale feature extraction and wavelet transform convolution, while combining the shared decoder for feature fusion. The high-performance dual feature extractor of H&E-IHC is trained by contrastive learning, which can effectively perform feature alignment of HE-IHC in high latitude space. At the same time, the trained feature encoder is used to enhance the features and adaptively adjust the loss in the HE section staining process to solve the problems related to unclear and asymmetric information. We have tested on different datasets and achieved excellent performance.Our code is available at https://github.com/babyinsunshine/CEFF

Problem

• Develop adaptive H&E-IHC information fusion staining
• Address information loss in H&E-to-IHC transformation
• Enhance feature alignment using contrastive learning

Innovation

• Adaptive H&E-IHC information fusion
• Multi-scale feature extraction
• Contrastive learning for feature alignment