MIMM-X：解构医学图像分析中的虚假相关性
MIMM-X: Disentangling Spurious Correlations for Medical Image Analysis
AI总结：
MIMM-X通过最小化多重虚假相关性的互信息，解构医学图像分析中的因果特征，提升模型在新环境下的泛化能力。
摘要：
深度学习模型在医疗任务中表现优异，但往往会出现虚假相关性，即捷径学习，导致在新环境中泛化能力差。特别是在医学成像中，多个虚假相关性可能同时存在，误分类可能产生严重后果。我们提出了MIMM-X框架，通过最小化多重虚假相关性的互信息来解构因果特征，使预测基于真实的因果关系而非数据集特定的捷径。我们在三个数据集（UK Biobank、NAKO、CheXpert）和两种成像模态（MRI和X射线）上评估了MIMM-X。结果表明，MIMM-X有效缓解了多重虚假相关性的捷径学习。
英文摘要：
Deep learning models can excel on medical tasks, yet often experience spurious correlations, known as shortcut learning, leading to poor generalization in new environments. Particularly in medical imaging, where multiple spurious correlations can coexist, misclassifications can have severe consequences. We propose MIMM-X, a framework that disentangles causal features from multiple spurious correlations by minimizing their mutual information. It enables predictions based on true underlying causal relationships rather than dataset-specific shortcuts. We evaluate MIMM-X on three datasets (UK Biobank, NAKO, CheXpert) across two imaging modalities (MRI and X-ray). Results demonstrate that MIMM-X effectively mitigates shortcut learning of multiple spurious correlations.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
