视频胶囊内镜中的罕见疾病分类
Rare Disease Classification in Video Capsule Endoscopy
查看机构详情
- University of South Dakota（南达科他大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
针对视频胶囊内镜中罕见疾病检测信号弱的问题，提出分层条件期望增益（HCE）解码方法，利用视频级疾病存在信息校正帧级预测，在多个数据集上显著提升罕见类召回率。
中文摘要 AI 辅助
视频胶囊内镜提供了一种非侵入性检查小肠的方式，然而，那些罕见但临床上重要的发现仅出现在不到1%的标注帧中，为检测提供的信号非常少。因此，尽管模型能够实现高整体性能，但它们几乎完全遗漏了这些罕见发现。我们将每一帧的决策基于该患者视频其余部分的证据，而非仅基于该帧自身的后验概率，从而提供了单个冻结帧永远无法包含的信息。基于期望增益仲裁（Expected Gain Arbitration），一种对混淆矩阵进行逐帧、宏F1最优的校正方法，我们进一步将其与视频疾病存在条件相结合，形成了我们的解码方法——分层条件期望增益（Hierarchical Conditional-Expected-gain, HCE）。我们在两个独立的VCE数据集（Kvasir-Capsule和GALAR）上，将HCE与静态时间平滑基线和最优传输（OT）帧级校正进行了评估。HCE在GALAR上相对于时间平滑基线显著提高了罕见类召回率（2.7--3.1倍；95%置信区间在两个范围内均排除零），并在Kvasir-Capsule上显示出类似的（尽管未经过统计确认）增益（最高3.4倍）。与OT相比，HCE在所有四种评估设置中均提高了罕见类召回率（+0.016至+0.029），并进一步提高了宏F1和马修斯相关系数（MCC）。
英文摘要
Video capsule endoscopy offers a non-invasive way to examine the small bowel, however, the rare but clinically important findings that appear only in less than 1\% of labelled frames, provide very less signal for detection. As a result, while models can achieve high overall performance, they almost completely miss these rare findings. We condition each frame's decision on evidence from the rest of that patient's video, not on the frame's own posterior alone, supplying information a single frozen frame can never contain. Building upon Expected Gain Arbitration, a per-frame, macro-F1-optimal correction over the confusion matrix, we further compose it with video-disease-presence conditional resulting in our decoding method Hierarchical Conditional-Expected-gain (HCE). We evaluate HCE on two independent VCE datasets, Kvasir-Capsule and GALAR, against a static temporal-smoothing baseline and an optimal-transport (OT) frame-level correction. HCE significantly improves rare-class recall over a temporal-smoothing baseline on GALAR (2.7--3.1x; 95\% CI excludes zero in both scopes) and shows a similar, though not statistically confirmed, gain on Kvasir-Capsule (up to 3.4x). Against OT, HCE still improves rare-class recall in all four evaluation settings (+0.016 to +0.029), and additionally improves macro-F1 and MCC.
