融合病历与胸片，提升异常检测准确率
Harnessing EHRs for Diffusion-based Anomaly Detection on Chest X-rays
- 引入图像-病历交叉注意力，融合临床信息增强生成
- 在CheXpert和MIMIC-CXR/IV上达到当前最佳性能
- 适合医学影像异常检测研究者参考
医学影像中的无监督异常检测对识别病理异常至关重要，且无需大量标注数据。然而，现有基于扩散模型的无监督方法仅依赖影像特征，难以区分正常解剖变异与病理异常。为此，我们提出Diff3M，一种多模态扩散框架，整合胸部X光片与结构化电子健康记录（EHRs），以提升异常检测能力。具体地，设计了新颖的图像-EHR交叉注意力模块，将临床上下文融入图像生成过程，增强模型对正常与异常特征的区分能力；同时，采用静态掩码策略，提升从异常中重建正常图像的效果。在CheXpert和MIMIC-CXR/IV上的广泛评估表明，Diff3M表现优于现有医学影像无监督异常检测方法。代码已公开于https://github.com/nth221/Diff3M。
原文摘要 · Abstract (English)
Unsupervised anomaly detection (UAD) in medical imaging is crucial for identifying pathological abnormalities without requiring extensive labeled data. However, existing diffusion-based UAD models rely solely on imaging features, limiting their ability to distinguish between normal anatomical variations and pathological anomalies. To address this, we propose Diff3M, a multi-modal diffusion-based framework that integrates chest X-rays and structured Electronic Health Records (EHRs) for enhanced anomaly detection. Specifically, we introduce a novel image-EHR cross-attention module to incorporate structured clinical context into the image generation process, improving the model's ability to differentiate normal from abnormal features. Additionally, we develop a static masking strategy to enhance the reconstruction of normal-like images from anomalies. Extensive evaluations on CheXpert and MIMIC-CXR/IV demonstrate that Diff3M achieves state-of-the-art performance, outperforming existing UAD methods in medical imaging. Our code is available at this http URL https://github.com/nth221/Diff3M
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。