English

LGD-Net: Latent-Guided Dual-Stream Network for HER2 Scoring with Task-Specific Domain Knowledge

Computer Vision and Pattern Recognition 2026-02-23 v1 Image and Video Processing

Abstract

It is a critical task to evalaute HER2 expression level accurately for breast cancer evaluation and targeted treatment therapy selection. However, the standard multi-step Immunohistochemistry (IHC) staining is resource-intensive, expensive, and time-consuming, which is also often unavailable in many areas. Consequently, predicting HER2 levels directly from H&E slides has emerged as a potential alternative solution. It has been shown to be effective to use virtual IHC images from H&E images for automatic HER2 scoring. However, the pixel-level virtual staining methods are computationally expensive and prone to reconstruction artifacts that can propagate diagnostic errors. To address these limitations, we propose the Latent-Guided Dual-Stream Network (LGD-Net), a novel framework that employes cross-modal feature hallucination instead of explicit pixel-level image generation. LGD-Net learns to map morphological H&E features directly to the molecular latent space, guided by a teacher IHC encoder during training. To ensure the hallucinated features capture clinically relevant phenotypes, we explicitly regularize the model training with task-specific domain knowledge, specifically nuclei distribution and membrane staining intensity, via lightweight auxiliary regularization tasks. Extensive experiments on the public BCI dataset demonstrate that LGD-Net achieves state-of-the-art performance, significantly outperforming baseline methods while enabling efficient inference using single-modality H&E inputs.

Keywords

Cite

@article{arxiv.2602.17793,
  title  = {LGD-Net: Latent-Guided Dual-Stream Network for HER2 Scoring with Task-Specific Domain Knowledge},
  author = {Peide Zhu and Linbin Lu and Zhiqin Chen and Xiong Chen},
  journal= {arXiv preprint arXiv:2602.17793},
  year   = {2026}
}
R2 v1 2026-07-01T10:43:34.382Z