Metraipolis › archive › dc_e857be75908f

MirrorNet: Can Medical Image Anonymization Really Protect Patient Identity? — arXiv:2608.05938

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.05938
License (as archived): CC0-1.0
Archived: 2026-10-31T10:36:24.186Z
SHA-256 of the archived text: deb2fab8af470a802d5413d502c807f66d6a046e7f6e8722c2d1cdbaf529709d
Stable URL of this page: https://metraipolis.com/documento/dc_e857be75908f

How to cite

MirrorNet: Can Medical Image Anonymization Really Protect Patient Identity? — arXiv:2608.05938 — arXiv (en), https://arxiv.org/abs/2608.05938 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e857be75908f (Metraipolis archive dc_e857be75908f, sha256 deb2fab8af470a802d5413d502c807f66d6a046e7f6e8722c2d1cdbaf529709d)

BibTeX

@misc{dce857be75908f,
  title = {MirrorNet: Can Medical Image Anonymization Really Protect Patient Identity? — arXiv:2608.05938},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.05938},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e857be75908f; sha256 deb2fab8af470a802d5413d502c807f66d6a046e7f6e8722c2d1cdbaf529709d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05938. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Medical images are routinely de-identified---names, dates, and other metadata removed---and then shared for research, teaching, and public benchmarks under the assumption that this renders them anonymous. Such de-identification protects the metadata but not the pixels, and---apart from scans that directly contain facial structures---whether the image content itself identifies the patient has received little scrutiny. We investigate this question by learning a cycle-consistent correspondence between a cross-sectional medical image and a non-medical, patient-identifying image, using a pair of coupled, cycle-consistent variational autoencoders. From a held-out scan, the model recovers a recognisable likeness of the patient (identity-region MAE = 0.163); conversely, it synthesises a scan from such an image. These results indicate that a de-identified medical scan remains identifying---it is, in effect, a photograph of the patient---and that imaging data should be governed as biometric data rather than as anonymisable records. To support reproducibility, the code and trained models are shared at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-07 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.05938 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-31T10:23:31.036Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).