通过跨窗知识蒸馏，挖掘肺部CT中隐含的病理特征。
Uncovering Latent Pathological Signatures in Pulmonary CT via Cross-Window Knowledge Distillation
- 学生编码器从教师模型学习跨窗临床先验
- 在三个数据集上提升AUC达10.1-16.5个百分点
- 适合需融合多窗肺CT信息的研究者
多窗CT成像可捕捉不同密度解剖结构间的互补病理信息，但现有深度学习方法仅在后期融合表示，遗漏跨密度交互。我们提出一种跨窗知识蒸馏框架，使学生编码器从在最具信息量窗口训练的教师模型中学习潜在临床先验。回顾性评估在三个队列上：COPD-CT-DF（n=719）、RSNA PE（n=1,433）和院内CTEPD数据集（n=161）。蒸馏使各窗AUC提升10.1-16.5个百分点（COPD-CT-DF：0.75-0.81 → 0.90-0.94；所有P<0.001），集成AUC达0.9960。类似增益见于RSNA PE（0.80-0.83 → 0.90-0.92）和CTEPD（AUC 0.7481 vs. 0.6264）。该方法内化了监督方法无法察觉的病理签名，为多窗肺CT分析提供通用解决方案。
原文摘要 · Abstract (English)
Multi-window CT imaging captures complementary pathological information across anatomical structures of differing densities, yet existing deep learning methods fuse representations only at later stages, missing cross-density interactions. We propose a cross-window knowledge distillation framework in which student encoders learn latent clinical priors from a teacher trained on the most informative window. Evaluated retrospectively on three cohorts - COPD-CT-DF (n=719), RSNA PE (n=1,433), and an in-house CTEPD dataset (n=161) - distillation improved per-window AUC by 10.1-16.5 percentage points on COPD-CT-DF (0.75-0.81 to 0.90-0.94; all P<0.001), with ensemble AUC reaching 0.9960. Similar gains were observed on RSNA PE (0.80-0.83 to 0.90-0.92) and CTEPD (AUC 0.7481 vs. 0.6264). Cross-window distillation internalises pathological signatures invisible to supervised approaches, offering a generalisable solution for multi-window pulmonary CT analysis.
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。