Metraipolis › archive › dc_4cd7e302a39a
Can saliency maps flag chest radiograph classifier errors at a new hospital? An external validation study — arXiv:2611.05011
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.05011
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T12:42:48.244Z
- SHA-256 of the archived text
d3d7e1202233c138c2485192a9361334f3d335772dc94e89e1ff41776b548ef9- Stable URL of this page
https://metraipolis.com/documento/dc_4cd7e302a39a
How to cite
Can saliency maps flag chest radiograph classifier errors at a new hospital? An external validation study — arXiv:2611.05011 — arXiv (en), https://arxiv.org/abs/2611.05011 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4cd7e302a39a (Metraipolis archive dc_4cd7e302a39a, sha256 d3d7e1202233c138c2485192a9361334f3d335772dc94e89e1ff41776b548ef9)
BibTeX
@misc{dc4cd7e302a39a,
title = {Can saliency maps flag chest radiograph classifier errors at a new hospital? An external validation study — arXiv:2611.05011},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.05011},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4cd7e302a39a; sha256 d3d7e1202233c138c2485192a9361334f3d335772dc94e89e1ff41776b548ef9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05011. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Saliency maps are offered as a safeguard for medical image classifiers, but whether they reveal individual errors at a new hospital remains largely untested. We trained a convolutional network and a vision transformer on NIH ChestX-ray14 and evaluated 303,768 decisions per model at the training hospital and at two external sites, in the USA and Vietnam, with radiologist-consensus labels. We scored each explanation by its anatomical enrichment, the area-normalized concentration of attribution in a segmented thoracic region, and tested whether it detects errors beyond the model's confidence. Grad-CAM localized findings less often than radiologists (hit rate at most 0.575 versus 0.888). Adding enrichment to a confidence-based error detector calibrated at the training hospital never raised its area under the curve by more than 0.011, and for the convolutional network the relation between enrichment and correctness reversed between hospitals, degrading the transferred detector. Model confidence, not saliency, should anchor deployment monitoring. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG); Image and Video Processing (eess.IV). Pagina del lavoro: https://arxiv.org/abs/2611.05011 — 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 2026-11-04T12:29:47.401Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).