Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.12562 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914560807534592 |
|---|---|
| author | Peng, Bo Xu, Wujian Wang, Kun Liao, Ximing Wang, Na Shi, Daqian Li, Tian Gao, Jing Thygesen, Johan Ji, Yingqun Wu, Honghan |
| author_facet | Peng, Bo Xu, Wujian Wang, Kun Liao, Ximing Wang, Na Shi, Daqian Li, Tian Gao, Jing Thygesen, Johan Ji, Yingqun Wu, Honghan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_12562 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Uncovering Latent Pathological Signatures in Pulmonary CT via Cross-Window Knowledge Distillation Peng, Bo Xu, Wujian Wang, Kun Liao, Ximing Wang, Na Shi, Daqian Li, Tian Gao, Jing Thygesen, Johan Ji, Yingqun Wu, Honghan Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition 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. |
| title | Uncovering Latent Pathological Signatures in Pulmonary CT via Cross-Window Knowledge Distillation |
| topic | Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.12562 |