[论文] Closing the Domain Gap in Biomedical Imaging by In-Context Control Sam...

## 论文概要 **研究领域**: ML **作者**: Ana Sanchez-Fernandez, Thomas Pinetz, Werner Zellinger **发布时间**: 2026-04-22 **arXiv**: [2604.20824](https://arxiv.org/abs/2604.20824) ## 中文摘要 生物医学成像的核心问题是批次效应：与感兴趣的生物学信号无关的系统性技术变异。这些批次效应严重损害实验可重复性，是导致深度学习系统在新实验批次上失败的主要原因，阻碍其在现实世界中的实际应用。尽管多年研究，尚无方法成功弥合深度学习模型的这一性能差距。我们提出CS-ARM-BN（Control-Stabilized Adaptive Risk Minimization via Batch Normalization），一种元学习适应方法，利用阴性对照样本。这类未扰动的参考图像在每个实验批次中按设计都存在，可作为适应的稳定上下文。我们在大规模JUMP-CP数据集的药物作用机制（MoA）分类任务上验证了新方法。标准ResNet的准确率从训练域的0.939±0.005下降到新实验批次数据的0.862±0.060。基础模型即使经过典型变异归一化也无法弥合这一差距。我们首次证明元学习方法通过达到0.935±0.018来弥合域差距。如果新实验批次表现出强域偏移（如在不同实验室生成），元学习方法可用对照样本稳定，而对照样本在生物医学实验中始终可用。我们的工作表明，生物医学成像数据中的批次效应可以通过有原则的上下文适应有效中和，这也使其实用且高效。 ## 原文摘要 The central problem in biomedical imaging are batch effects: systematic technical variations unrelated to the biological signal of interest. These batch effects critically undermine experimental reproducibility and are the primary cause of failure of deep learning systems on new experimental batches, preventing their practical use in the real world. Despite years of research, no method has succeeded in closing this performance gap for deep learning models. We propose Control-Stabilized Adaptive Risk Minimization via Batch Normalization (CS-ARM-BN), a meta-learning adaptation method that exploits negative control samples. Such unperturbed reference images are present in every experimental batch by design and serve as stable context for adaptation. We validate our novel method on Mechanism-o... --- *自动采集于 2026-04-24* #论文 #arXiv #ML #小凯