对抗性自监督学习提升空间蛋白质组学图像的生物意义表示
AdvDINO: Domain-Adversarial Self-Supervised Representation Learning for Spatial Proteomics
- 在DINOv2中引入梯度反转层，实现域不变特征学习
- 在超546万张图像上识别出具有预后意义的蛋白表型簇
- 适用于肺癌、乳腺癌等医学影像，缓解批次效应干扰
自监督学习（SSL）在无需人工标注的情况下学习视觉表征方面表现强大。然而，标准SSL方法对域偏移——数据源间的系统性差异——的鲁棒性仍不明确，这在生物医学成像中尤为关键，因批次效应可能掩盖真实生物学信号。我们提出AdvDINO，一种域对抗性自监督框架，将梯度反转层集成到DINOv2架构中，以促进域不变特征学习。该方法应用于六通道多重免疫荧光（mIF）全切片图像的真实患者队列，有效缓解了切片特异性偏差，学习到比非对抗基线更稳健、更具生物学意义的表征。在超过546万张mIF图像块上，模型识别出具有不同蛋白组特征和预后意义的表型簇，并通过基于注意力的多实例学习实现优异的生存预测性能。其增强的鲁棒性也扩展至乳腺癌队列。尽管在mIF数据上验证，AdvDINO广泛适用于其他存在域偏移的医学影像领域。
原文摘要 · Abstract (English)
Self-supervised learning (SSL) has emerged as a powerful approach for learning visual representations without manual annotations. However, the robustness of standard SSL methods to domain shift -- systematic differences across data sources -- remains uncertain, posing an especially critical challenge in biomedical imaging where batch effects can obscure true biological signals. We present AdvDINO, a domain-adversarial SSL framework that integrates a gradient reversal layer into the DINOv2 architecture to promote domain-invariant feature learning. Applied to a real-world cohort of six-channel multiplex immunofluorescence (mIF) whole slide images from lung cancer patients, AdvDINO mitigates slide-specific biases to learn more robust and biologically meaningful representations than non-adversarial baselines. Across more than 5.46 million mIF image tiles, the model uncovers phenotype clusters with differing proteomic profiles and prognostic significance, and enables strong survival prediction performance via attention-based multiple instance learning. The improved robustness also extends to a breast cancer cohort. While demonstrated on mIF data, AdvDINO is broadly applicable to other medical imaging domains, where domain shift is a common challenge.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。