迈向可靠的染色转移：基于多模态专家指导评估的迭代数据-模型协同优化框架
Towards Reliable Stain Transfer: An Iterative Data-Model Co-Optimization Framework Based on Multimodal Expert-Guided Assessment
浏览论文内容
中文总结 AI 辅助
研究针对染色转移建模难题，提出DMCoStain框架，通过迭代优化数据与模型能力提升准确性和可解释性。基于IPE视觉语言模型构建MEGFS策略并创建ImmunoInstruction数据集，实验证明该框架达最优精度，其范式有实用价值，MEGFS可作评估工具。
中文摘要 AI 辅助
组织病理学检查主要依靠苏木精和伊红（H&E）以及免疫组织化学（IHC）染色。虽然IHC提供关键分子信息，但成本高且需专业知识。染色转移可通过计算从H&E图像生成IHC，但在像素未对齐监督下对异质生物标志物进行统一且可解释的建模仍面临挑战。我们提出了DMCoStain，一种用于染色转移的新型数据-模型协同优化框架。它迭代地共同优化训练数据和模型能力，提高染色准确性以及病理和结构一致性方面的可解释性。为以临床有意义的方式优化训练数据，它纳入了基于模拟病理学家推理的开创性IHC阳性表达（IPE）视觉语言模型（VLM）构建的多模态专家指导精细选择（MEGFS）策略。为支持MEGFS，我们构建了ImmunoInstruction，首个拥有15万个VQA样本的大规模IPE指令跟随数据集。对多种组织和生物标志物的广泛实验表明，DMCoStain实现了当前最优（SOTA）准确性。该范式具有很强的实用价值，MEGFS还可作为未来模型开发的专业评估工具。数据集、代码及更多细节见此https链接。
英文摘要
Histopathological examination primarily relies on hematoxylin and eosin (H&E) and immunohistochemistry (IHC) staining. Although IHC provides critical molecular information, it is costly and requires specialized expertise. Stain transfer provides an efficient alternative by computationally generating IHC from H&E images, but remains challenged by unified and interpretable modeling for heterogeneous biomarkers under pixel-unaligned supervision. We propose DMCoStain, a novel Data-Model Co-optimization framework for Stain transfer. It iteratively co-refines training data and model capability, improving staining accuracy and interpretability in both pathological and structural consistency. To refine training data in a clinically meaningful manner, it incorporates the Multimodal Expert-Guided Finer Selection (MEGFS) strategy, built upon a pioneering IHC-positive-expression (IPE) vision-language model (VLM) that emulates pathologist reasoning. To support MEGFS, we construct ImmunoInstruction, the first large-scale IPE instruction-following dataset with 150K VQA samples. Extensive experiments on multiple tissues and biomarkers demonstrate that DMCoStain achieves state-of-the-art (SOTA) accuracy. This paradigm offers strong practical value, and MEGFS also functions as a specialized evaluation tool for future model development. Dataset, code, and more details are in https://github.com/SikangSHU/DMCoStain.
发表机构
- East China Normal University（华东师范大学）
- Ruijin Hospital, Shanghai Jiao Tong University School of Medicine（上海交通大学医学院附属瑞金医院）
- Hangzhou Hyperspectral Imaging Technology Co., Ltd.（杭州高光谱成像技术有限公司）
- Fudan University Shanghai Cancer Center（复旦大学附属肿瘤医院）
机构由 AI 辅助整理，请以论文原文为准。
