CoChat
Sign Up
Cited by 8COVID-19 diagnosis using AIAdvanced Neural Network ApplicationsAI in cancer detection

The DOI matches its Crossref record; 3 of the 3 references it cites were found in a registry or an index.

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

Open access
Müller, Philip, Meissen, Felix, Kaissis, Georgios, Rueckert, Daniel·2025·IEEE Transactions on Medical Imaging
DOI 10.1109/tmi.2024.3435015
Open this paper in CoChat — free account

Abstract

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.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/tmi.2024.3435015

Cites

3 of 3 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 42

  1. 10.1609/aaai.v33i01.3301590 · DOI · Paper page
  2. 10.1109/CVPR.2017.369 · DOI · Paper page
  3. 10.1109/CVPR.2015.7298668 · DOI · Paper page
  4. 10.1109/CVPR.2016.311 · DOI · Paper page
  5. 10.1007/978-3-319-46454-1_22 · DOI
  6. Ramon, Multi instance neural networks, 53
  7. 10.1007/s11263-013-0620-5 · DOI · Paper page
  8. 10.1007/978-3-319-10602-1_26 · DOI · Paper page
  9. Rajpurkar, 2017, CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning, arXiv:1711.05225 · arXiv
  10. 10.1007/978-3-319-46723-8_28 · DOI
  11. 10.1145/3233547.3233573 · DOI
  12. 10.1109/ACCESS.2020.3005510 · DOI
  13. 10.1016/j.neucom.2022.01.095 · DOI
  14. Everingham, 2024, The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Results
  15. 2024, The PASCAL Visual Object Classes Challenge 2012 (VOC2012) Results
  16. 10.1007/978-3-319-10602-1_48 · DOI · Paper page
  17. 10.1007/s11263-015-0816-y · DOI · Paper page
  18. 10.1109/CVPR.2017.326 · DOI
  19. 10.1109/TPAMI.2018.2876304 · DOI · Paper page
  20. 10.1109/CVPR42600.2020.01301 · DOI
  21. 10.1109/TPAMI.2015.2389824 · DOI · Paper page
  22. 10.1109/CVPR42600.2020.01061 · DOI
  23. 10.1007/978-3-031-20077-9_13 · DOI
  24. 10.1109/CVPR.2016.319 · DOI · Paper page
  25. 10.1109/CVPR.2015.7298780 · DOI
  26. 10.1109/CVPR.2016.513 · DOI
  27. 10.1109/ICCV.2017.74 · DOI · Paper page
  28. 10.1109/CVPR.2018.00144 · DOI
  29. 10.1007/978-3-031-16443-9_63 · DOI
  30. 10.1007/978-3-030-00919-9_29 · DOI
  31. 10.1109/CVPR.2017.243 · DOI · Paper page
  32. 10.1007/s10032-020-00360-2 · DOI
  33. 10.5555/3295222.3295349 · DOI
  34. 10.1109/ICCV.2015.169 · DOI · Paper page
  35. Maron, A framework for multiple-instance learning, 570
  36. 10.1109/CVPR.2014.312 · DOI
  37. Babenko, Simultaneous learning and alignment: Multi-instance and multi-pose learning
  38. Khosla, Supervised contrastive learning, 33, 18661
  39. Chen, A simple framework for contrastive learning of visual representations, 1597
  40. Meissen, 2023, Robust detection outcome: A metric for pathology detection in medical images, arXiv:2303.01920 · arXiv
  41. Paszke, PyTorch: An imperative style, high-performance deep learning library, 32
  42. Loshchilov, Decoupled weight decay regularization

Cited by

Sorted by citations

Metadata

Authors
Müller, P., Meissen, F., Kaissis, G., & Rueckert, D.
Affiliations
  • Technical University of Munich (DE)
  • Helmholtz Munich (DE)
  • Imperial College London (GB)
ORCID
Venue
IEEE Transactions on Medical Imaging
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
2025-01
Volume
44
Issue
1
Pages
221-231
ISSN
  • 0278-0062 (ISSN-L)
  • 1558-254X
DOI
10.1109/tmi.2024.3435015
OpenAlex
W4401070680
Cited by
8
References
3
Citations by year
2026: 2 · 2025: 5
FWCI
2.04
Type
journal article
Language
en
Open access
Licence
Locations
  • IEEE Transactions on Medical Imaging (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • mediaTUM (Technical University of Munich) (submittedVersion)
  • PuSH - Publication Server of Helmholtz Zentrum München (submittedVersion)
  • arXiv (Cornell University)
Funders
ERC Project Deep4MI (884622)
Topics
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
Keywords
weakly supervised object detection, chest radiography, disease localization, medical image classification, multiple instance learning
MeSH terms
Supervised Machine Learning, Algorithms, Humans, Image Processing, Computer-Assisted, Lung, Radiographic Image Interpretation, Computer-Assisted, Radiography, Thoracic, Neural Networks, Computer

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review