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.
@article{arxiv.2605.12562,
title = {Uncovering Latent Pathological Signatures in Pulmonary CT via Cross-Window Knowledge Distillation},
author = {Bo Peng and Wujian Xu and Kun Wang and Ximing Liao and Na Wang and Daqian Shi and Tian Li and Jing Gao and Johan Thygesen and Yingqun Ji and Honghan Wu},
journal= {arXiv preprint arXiv:2605.12562},
year = {2026}
}