EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Mislabel Detection in Multi-Label Chest X-Rays via Prototype-Weighted Neighborhood Consistency in CoAtNet Embedding Space

Ariel Gamboa, Mauricio Araya, Camilo Sotomayor

Applied Sciences · 22.04.2026 · 16 (9) · 4067

DOI: 10.3390/app16094067 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 22.04.2026. Verfasst von Ariel Gamboa und 2 weiteren Autor:innen. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: ANID AC3E.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large-scale chest X-ray (CXR) datasets often rely on report-derived or weak labels, introducing missing and incorrect annotations that can degrade downstream models and limit trust. We study training-free mislabel detection in multi-label CXRs by scoring neighborhood label consistency in a fixed embedding space. Using the NIH Chest X-ray Kaggle sample (5606 CXRs), we extract intermediate CoAtNet features and obtain 64-dimensional embeddings with a frozen CoAtNet backbone and a lightweight refinement head. On top of these embeddings, we compare kNN consistency baselines with distance weighting and label-set similarity against LPV-DW-CS, clustered prototype voting weighted by distance and cluster support. We evaluate three synthetic label-noise regimes with review budgets matched to the corruption rate: random single-label (5% and 20%), boundary-noise (20% corruption within the lowest-density 20% subset), and disjoint-label replacement (20% within that subset). LPV-DW-CS yields the highest downstream macro-AUROC after filtering top-ranked samples (up to 0.8860), while kNN variants achieve higher Recall@budget at the same review rates (up to 99.44%). An image-only expert Likert review of top-ranked real samples finds substantial label-set inconsistencies (54.1% for LPV-DW-CS-280-A; 60.5% for KNN-DW-LSS), supporting neighborhood-consistency ranking as a practical, training-free tool for targeted dataset auditing.

Förderung

  • ANID AC3E

Zitieren

APA

Gamboa, A., Araya, M., & Sotomayor, C. (2026). Mislabel Detection in Multi-Label Chest X-Rays via Prototype-Weighted Neighborhood Consistency in CoAtNet Embedding Space. Applied Sciences, 16(9), 4067. https://doi.org/10.3390/app16094067

Vancouver

Gamboa A, Araya M, Sotomayor C. Mislabel Detection in Multi-Label Chest X-Rays via Prototype-Weighted Neighborhood Consistency in CoAtNet Embedding Space. Applied Sciences. 2026;16(9):4067. doi:10.3390/app16094067

BibTeX
@article{Gamboa2026,
  author = {Gamboa, Ariel and Araya, Mauricio and Sotomayor, Camilo},
  title = {Mislabel Detection in Multi-Label Chest X-Rays via Prototype-Weighted Neighborhood Consistency in CoAtNet Embedding Space},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {9},
  pages = {4067},
  doi = {10.3390/app16094067}
}
Kostenlos starten →