EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Weakly Supervised Object Detection in Chest X-Rays With Differentiable ROI Proposal Networks and Soft ROI Pooling

Philip Muller, Felix Meissen, Georgios Kaissis, Daniel Rueckert

IEEE Transactions on Medical Imaging · 2025

DOI: 10.1109/tmi.2024.3435015 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2025. Verfasst von Philip Muller und 3 weiteren Autor:innen. Hauptthemen laut MEDLINE: Radiography, Thoracic, Supervised Machine Learning, Radiographic Image Interpretation, Computer-Assisted, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: ERC Project Deep4MI.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Weakly supervised object detection (WSup-OD) increases the usefulness and interpretability of image classification algorithms without requiring additional supervision. The successes of multiple instance learning in this task for natural images, however, do not translate well to medical images due to the very different characteristics of their objects (i.e. pathologies). In this work, we propose Weakly Supervised ROI Proposal Networks (WSRPN), a new method for generating bounding box proposals on the fly using a specialized region of interest-attention (ROI-attention) module. WSRPN integrates well with classic backbone-head classification algorithms and is end-to-end trainable with only image-label supervision. We experimentally demonstrate that our new method outperforms existing methods in the challenging task of disease localization in chest X-ray images. Code: https://github.com/philip-mueller/wsrpn.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Radiography, ThoracicSupervised Machine LearningRadiographic Image Interpretation, Computer-AssistedImage Processing, Computer-AssistedHumansAlgorithmsLung

Förderung

  • ERC Project Deep4MI

Zitieren

APA

Muller, P., Meissen, F., Kaissis, G., & Rueckert, D. (2025). Weakly Supervised Object Detection in Chest X-Rays With Differentiable ROI Proposal Networks and Soft ROI Pooling. IEEE Transactions on Medical Imaging. https://doi.org/10.1109/tmi.2024.3435015

Vancouver

Muller P, Meissen F, Kaissis G, Rueckert D. Weakly Supervised Object Detection in Chest X-Rays With Differentiable ROI Proposal Networks and Soft ROI Pooling. IEEE Transactions on Medical Imaging. 2025. doi:10.1109/tmi.2024.3435015

BibTeX
@article{Muller2025,
  author = {Muller, Philip and Meissen, Felix and Kaissis, Georgios and Rueckert, Daniel},
  title = {Weakly Supervised Object Detection in Chest X-Rays With Differentiable ROI Proposal Networks and Soft ROI Pooling},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2025},
  doi = {10.1109/tmi.2024.3435015}
}
Kostenlos starten →