CoChat

Learning to detect chest radiographs containing pulmonary lesions using visual attention networks

Open access
Pesce, Emanuele, Joseph Withey, Samuel, Ypsilantis, Petros-Pavlos, Bakewell, Robert, Goh, Vicky, Montana, Giovanni · 2019 · Medical Image Analysis
Cited by 126

The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.

Abstract

Machine learning approaches hold great potential for the automated detection of lung nodules on chest radiographs, but training algorithms requires very large amounts of manually annotated radiographs, which are difficult to obtain. The increasing availability of PACS (Picture Archiving and Communication System), is laying the technological foundations needed to make available large volumes of clinical data and images from hospital archives. Binary labels indicating whether a radiograph contains a pulmonary lesion can be extracted at scale, using natural language processing algorithms. In this study, we propose two novel neural networks for the detection of chest radiographs containing pulmonary lesions. Both architectures make use of a large number of weakly-labelled images combined with a smaller number of manually annotated x-rays. The annotated lesions are used during training to deliver a type of visual attention feedback informing the networks about their lesion localisation performance. The first architecture extracts saliency maps from high-level convolutional layers and compares the inferred position of a lesion against the true position when this information is available; a localisation error is then back-propagated along with the softmax classification error. The second approach consists of a recurrent attention model that learns to observe a short sequence of smaller image portions through reinforcement learning; the reward function penalises the exploration of areas, within an image, that are unlikely to contain nodules. Using a repository of over 430,000 historical chest radiographs, we present and discuss the proposed methods over related architectures that use either weakly-labelled or annotated images only.

Overview

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

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

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

Integrity signals

Registry

Crossref · 10.1016/j.media.2018.12.007

Cites

19 of 19 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.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

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

Research works printed but not linked: 31

  1. American Cancer Society, 1999. Key statistics for lung cancer. https://www.cancer.org/cancer/non-small-cell-lung-cancer/about/key-statistics.html, arXiv:1711.00583. · arXiv
  2. Anderson, 1994, The efficacy of double reading mammograms in breast screening, Clin. Radiol., 49, 248, 10.1016/S0009-9260(05)81850-1 · DOI
  3. Brady, 2017, Error and discrepancy in radiology: inevitable or avoidable?, Insights Imag., 8, 171, 10.1007/s13244-016-0534-1 · DOI · Paper page
  4. Chapman, 2013, Extending the negex lexicon for multiple languages, Stud. Health Technol. Inform., 677
  5. Cichy, 2014, Resolving human object recognition in space and time, Nat. Neurosci., 17, 455, 10.1038/nn.3635 · DOI · Paper page
  6. Cornia, 2016, A deep multi-level network for saliency prediction.
  7. Denil, 2011, Learning where to attend with deep architectures for image tracking.
  8. Elmore, 1994, Variability in radiologists’ interpretations of mammograms, N. Engl. J. Med., 331, 1493, 10.1056/NEJM199412013312206 · DOI · Paper page
  9. Everingham, 2010, The pascal visual object classes (VOC) challenge, Int. J. Comput. Vis., 88, 303, 10.1007/s11263-009-0275-4 · DOI · Paper page
  10. Ferlay, 2013, Cancer incidence and mortality patterns in europe: estimates for 40 countries in 2012, Eur. J. Cancer, 49, 1374, 10.1016/j.ejca.2012.12.027 · DOI · Paper page
  11. Forrest, 1981, Radiologic errors in patients with lung cancer, Western J. Med., 134, 485
  12. Guan, 2017, Who said what: modeling individual labelers improves classification.
  13. Hansell, 2008, Fleischner society: glossary of terms for thoracic imaging, Radiology, 246, 697, 10.1148/radiol.2462070712 · DOI · Paper page
  14. Langlotz, 2006, Radlex: a new method for indexing online educational materials, Radiographics, 26, 1595, 10.1148/rg.266065168 · DOI · Paper page
  15. Langlotz, 2006, Radlex: a new method for indexing online educational materials, Radiographics, 26, 1595, 10.1148/rg.266065168 · DOI · Paper page
  16. Lin, 2014, Microsoft coco: common objects in context.
  17. Manning, 2014, The stanford coreNLP natural language processing toolkit, In Association for Computational Linguistics (ACL) System Demonstrations, 55, 10.3115/v1/P14-5010 · DOI · Paper page
  18. Marbach, 2003, Approximate gradient methods in policy-space optimization of markov reward processes, Discret. Event Dyn. Syst., 13, 111, 10.1023/A:1022145020786 · DOI
  19. Mehrabi, 2015, DEEPEN: a negation detection system for clinical text incorporating dependency relation into negex, J. Biomed. Inform., 54, 213, 10.1016/j.jbi.2015.02.010 · DOI · Paper page
  20. Moore, 2011, Sensitivity and specificity of a cad solution for lung nodule detection on chest radiograph with cta correlation, J. Digit. Imag., 24, 405, 10.1007/s10278-010-9284-7 · doi:10.1007/s10278-010-9284-7
  21. Northcutt, 2017, Learning with confident examples: rank pruning for robust classification with noisy labels, ArXiv e-prints
  22. Quekel, 1999, Miss rate of lung cancer on the chest radiograph in clinical practice, Chest, 115, 720, 10.1378/chest.115.3.720 · DOI · Paper page
  23. Ranzato, 2014, On learning where to look.
  24. Rensink, 2000, The dynamic representation of scenes, Vis. cogn., 7, 17, 10.1080/135062800394667 · DOI · Paper page
  25. Sermanet, 2014, Attention for fine-grained categorization.
  26. Shin, 2016, Learning to read chest x-rays: recurrent neural cascade model for automated image annotation.
  27. Simonyan, 2014, Very deep convolutional networks for large-scale image recognition.
  28. Szucs-Farkas, 2013, Comparison of dual-energy subtraction and electronic bone suppression combined with computer-aided detection on chest radiographs: effect on human observers’ performance in nodule detection, Am. Jo. Roentgenol., 200, 1006, 10.2214/AJR.12.8877 · doi:10.2214/ajr.12.8877
  29. Yao, 2017, Deep learning from noisy image labels with quality embedding
  30. Zamir, 2016, Fedback networks.
  31. Zhou, 2016, Learning deep features for discriminative localization, CVPR

Cited by

See all 126

Metadata

Authors
Pesce, E., Joseph Withey, S., Ypsilantis, P.-P., Bakewell, R., Goh, V., & Montana, G.
Venue
Medical Image Analysis
Publisher
Elsevier BV
Published
2019-04
Volume
53
Pages
26-38
ISSN
1361-8415 (ISSN-L)
OpenAlex
W2775469102
Cited by
126
References
19
FWCI
8.52
Type
journal article
Language
en
Open access
hybrid
Licence
www.elsevier.com/tdm/userlicense/1.0/ (tdm)
Funders
  • National Institute for Health Research
  • Biomedical Research Centre
  • Guy’s & St Thomas’ NHS Foundation Trust
  • King’s College Hospital NHS Foundation Trust
  • King’s College London/University College
  • Cancer Research UK
  • Engineering and Physical Sciences Research Council
  • Medical Research Council

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

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.