用扩散模型修复医学图像，减少模型对虚假特征的依赖。
MaskMedPaint: Masked Medical Image Inpainting with Diffusion Models for Mitigation of Spurious Correlations
- 用文本驱动扩散模型修复关键区域外的图像内容
- 在少量目标域数据下提升跨域泛化能力
- 适合需可解释性的医疗图像分类任务
与类别标签相关的虚假特征会导致图像分类器依赖无法泛化的捷径。这在医疗领域尤为严重，因为有偏模型在不同医院或系统间应用时会失效。此时，基于数据的方法更受青睐，因临床医生可直接验证修改后的图像。尽管去噪扩散概率模型在自然图像上表现良好，但在医疗场景中不实用，因难以描述虚假医学特征。为此，我们提出掩码医学图像修复（MaskMedPaint），利用文本到图像扩散模型，通过修复关键分类区域外的区域，使训练图像匹配目标域分布。实验表明，在仅有限未标注目标域图像的前提下，MaskMedPaint在自然图像（Waterbirds、iWildCam）和医学数据集（ISIC 2018、Chest X-ray）上均显著提升了跨域泛化性能。
原文摘要 · Abstract (English)
Spurious features associated with class labels can lead image classifiers to rely on shortcuts that don't generalize well to new domains. This is especially problematic in medical settings, where biased models fail when applied to different hospitals or systems. In such cases, data-driven methods to reduce spurious correlations are preferred, as clinicians can directly validate the modified images. While Denoising Diffusion Probabilistic Models (Diffusion Models) show promise for natural images, they are impractical for medical use due to the difficulty of describing spurious medical features. To address this, we propose Masked Medical Image Inpainting (MaskMedPaint), which uses text-to-image diffusion models to augment training images by inpainting areas outside key classification regions to match the target domain. We demonstrate that MaskMedPaint enhances generalization to target domains across both natural (Waterbirds, iWildCam) and medical (ISIC 2018, Chest X-ray) datasets, given limited unlabeled target images.
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。