Metraipolis › archive › dc_d8f8efe8d8ec
Mitigating Spatial Blindness in Multiple Instance Learning for Whole-Slide Image Analysis — arXiv:2611.04337
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.04337
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T19:03:10.375Z
- SHA-256 of the archived text
922c52fd2a1585400b54f95ec2b4e97bd4ac4ac80857c5d0ac9a1ab365833968- Stable URL of this page
https://metraipolis.com/documento/dc_d8f8efe8d8ec
How to cite
Mitigating Spatial Blindness in Multiple Instance Learning for Whole-Slide Image Analysis — arXiv:2611.04337 — arXiv (en), https://arxiv.org/abs/2611.04337 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d8f8efe8d8ec (Metraipolis archive dc_d8f8efe8d8ec, sha256 922c52fd2a1585400b54f95ec2b4e97bd4ac4ac80857c5d0ac9a1ab365833968)
BibTeX
@misc{dcd8f8efe8d8ec,
title = {Mitigating Spatial Blindness in Multiple Instance Learning for Whole-Slide Image Analysis — arXiv:2611.04337},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.04337},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d8f8efe8d8ec; sha256 922c52fd2a1585400b54f95ec2b4e97bd4ac4ac80857c5d0ac9a1ab365833968},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04337. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Whole-slide multiple instance learning can predict diagnostic and prognostic labels from tissue composition while making limited use of spatial organization. Providing coordinates or a graph does not establish that a trained predictor uses tissue neighborhoods. We examine this gap as spatial blindness and hypothesize that early composition learning can weaken supervision of a spatial pathway. We introduce ResTopoMIL, which freezes a learned composition predictor before training a graph residual with a coordinate-reassignment separation loss. Permutation and information analyses distinguish architectural symmetry, available spatial information and the proposed optimization mechanism. On a controlled spatial task, staged training achieved an AUC of 0.987, compared with 0.684 for the same jointly trained architecture. Performance remained high on a composition-only task. Across four WSI classification and five survival benchmarks, ResTopoMIL improved mean primary performance on eight tasks; four improvements remained significant after Holm-adjusted patient-cluster tests. Coordinate-free residual controls recovered some of the gains, while neighborhood destruction reduced performance close to the composition anchor. Neighborhood-preserving transformations and anatomy-stratified perturbations further characterized this dependence. These results support fixing a composition reference as a practical intervention for learning spatial corrections, while retaining additional appearance capacity as a contributing explanation. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2611.04337 — 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 2026-11-04T18:39:37.491Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).