Metraipolis › archive › dc_94f487e2b077
Subgroup performance analysis of adaptation strategies for chest X-ray foundation models — arXiv:2608.19078
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.19078
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T14:06:34.917Z
- SHA-256 of the archived text
db8610db3b5546c33a9c6413ce2e8d55bf817bfbce9836b7867736f796dccc61- Stable URL of this page
https://metraipolis.com/documento/dc_94f487e2b077
How to cite
Subgroup performance analysis of adaptation strategies for chest X-ray foundation models — arXiv:2608.19078 — arXiv (en), https://arxiv.org/abs/2608.19078 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_94f487e2b077 (Metraipolis archive dc_94f487e2b077, sha256 db8610db3b5546c33a9c6413ce2e8d55bf817bfbce9836b7867736f796dccc61)
BibTeX
@misc{dc94f487e2b077,
title = {Subgroup performance analysis of adaptation strategies for chest X-ray foundation models — arXiv:2608.19078},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.19078},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_94f487e2b077; sha256 db8610db3b5546c33a9c6413ce2e8d55bf817bfbce9836b7867736f796dccc61},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19078. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Foundation models are increasingly adapted for downstream medical imaging tasks, yet the influence of the chosen adaptation strategy on subgroup fairness remains poorly understood. We investigate how three parameter-efficient adaptation techniques, including linear heads on the raw CLS token, an MLP, and an attention-pooling module over multi-layer patch features, affect both pathology classification performance and subgroup disparities when applied to the frozen Rad-DINO chest X-ray encoder. Using MIMIC-CXR, we evaluate eight pathologies across race, sex, and imaging-view subgroups on a prevalence-preserving, demographically balanced test set, and additionally probe how strongly each adapter encodes protected attributes. We find that attention pooling achieves the strongest overall discriminative performance and encodes attributes, particularly race, most strongly, but that improved overall performance does not consistently reduce subgroup disparities. Notably, stronger attribute encoding did not correspond to larger disparities: early network layers encoded race most weakly yet produced the largest subgroup performance gaps. Exploring different attention-pooling layer combinations further revealed no consistent relationship between the layers pooled, attribute encoding strength, and subgroup fairness. Our results indicate that richer, more expressive representations can improve accuracy while leaving fairness implications task-dependent and unpredictable, which must be assessed directly and per-task rather than inferred from encoding strength or overall performance alone. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-20 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.19078 — 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 2026-10-18T14:00:22.596Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).