Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2604.07128 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917391926034432 |
|---|---|
| author | Liu, Chenhao Wen, Zelin Tong, Yan Zhu, Junjie Tian, Xinyu Liu, Yuchi Gupta, Ashu Islam, Syed M. S. Gedeon, Tom Yao, Yue |
| author_facet | Liu, Chenhao Wen, Zelin Tong, Yan Zhu, Junjie Tian, Xinyu Liu, Yuchi Gupta, Ashu Islam, Syed M. S. Gedeon, Tom Yao, Yue |
| contents | Large-scale radiology data are critical for developing robust medical AI systems. However, sharing such data across hospitals remains heavily constrained by privacy concerns. Existing de-identification research in radiology mainly focus on removing identifiable information to enable compliant data release. Yet whether de-identified radiology data can still preserve sufficient utility for large-scale vision-language model training and cross-hospital transfer remains underexplored. In this paper, we introduce a utility-preserving de-identification pipeline (UPDP) for cross-hospital radiology data sharing. Specifically, we compile a blacklist of privacy-sensitive terms and a whitelist of pathology-related terms. For radiology images, we use a generative filtering mechanism that synthesis a privacy-filtered and pathology-reserved counterparts of the original images. These synthetic image counterparts, together with ID-filtered reports, can then be securely shared across hospitals for downstream model development and evaluation. Experiments on public chest X-ray benchmarks demonstrate that our method effectively removes privacy-sensitive information while preserving diagnostically relevant pathology cues. Models trained on the de-identified data maintain competitive diagnostic accuracy compared with those trained on the original data, while exhibiting a marked decline in identity-related accuracy, confirming effective privacy protection. In the cross-hospital setting, we further show that de-identified data can be combined with local data to yield better performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_07128 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Utility-preserving De-identification Pipeline for Cross-hospital Radiology Data Sharing Liu, Chenhao Wen, Zelin Tong, Yan Zhu, Junjie Tian, Xinyu Liu, Yuchi Gupta, Ashu Islam, Syed M. S. Gedeon, Tom Yao, Yue Computer Vision and Pattern Recognition Large-scale radiology data are critical for developing robust medical AI systems. However, sharing such data across hospitals remains heavily constrained by privacy concerns. Existing de-identification research in radiology mainly focus on removing identifiable information to enable compliant data release. Yet whether de-identified radiology data can still preserve sufficient utility for large-scale vision-language model training and cross-hospital transfer remains underexplored. In this paper, we introduce a utility-preserving de-identification pipeline (UPDP) for cross-hospital radiology data sharing. Specifically, we compile a blacklist of privacy-sensitive terms and a whitelist of pathology-related terms. For radiology images, we use a generative filtering mechanism that synthesis a privacy-filtered and pathology-reserved counterparts of the original images. These synthetic image counterparts, together with ID-filtered reports, can then be securely shared across hospitals for downstream model development and evaluation. Experiments on public chest X-ray benchmarks demonstrate that our method effectively removes privacy-sensitive information while preserving diagnostically relevant pathology cues. Models trained on the de-identified data maintain competitive diagnostic accuracy compared with those trained on the original data, while exhibiting a marked decline in identity-related accuracy, confirming effective privacy protection. In the cross-hospital setting, we further show that de-identified data can be combined with local data to yield better performance. |
| title | A Utility-preserving De-identification Pipeline for Cross-hospital Radiology Data Sharing |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.07128 |