用统计信息和少量像素标注训练医学图像分割模型
Learning to Segment using Summary Statistics and Weak Supervision
- 结合图像重建、统计匹配与弱监督重叠损失
- 仅用区域面积等统计量+少量像素标注即达较好效果
- 适合标注成本高的医学图像分割任务
医学专家常手动分割图像以获取诊断统计量，但标注后会丢弃。我们旨在训练分割模型减轻这一负担，仅依赖保留的统计量（如区域面积）。实验证明仅靠统计量不足以完成任务，但加入少量感兴趣区域内的像素作为弱监督信号可显著提升性能。采用一种新型损失函数，包含图像重建质量、统计量匹配以及预测前景与弱监督信号的重叠项。在标准图像、超声（乳腺癌）和计算机断层扫描（CT，肾肿瘤）数据集上的实验验证了该方法的有效性与潜力。
原文摘要 · Abstract (English)
Medical experts often manually segment images to obtain diagnostic statistics and discard the resulting annotations. We aim to train segmentation models to alleviate this burden, but constrained to the retained summary statistics (e.g., the area of the annotated region). Empirical results suggest that statistics alone are insufficient for this task, but adding weak information in the form of a few pixels within the area of interest significantly improves performance. We use a novel loss function that combines terms for image reconstruction quality, matching to summary statistics, and overlap between the predicted foreground and the weak supervisory signal. Experiments on standard image, ultrasound (breast cancer), and Computed Tomography (CT) scan (kidney tumors) data demonstrate the utility and potential of the approach.
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。