CoChat
Sign Up
Cited by 218

Weakly-supervised learning for lung carcinoma classification using deep learning

Kanavati, Fahdi, Toyokawa, Gouji, Momosaki, Seiya, Rambeau, Michael, Kozuma, Yuka, Shoji, Fumihiro, Yamazaki, Koji, Takeo, Sadanori +2 more·2020·Scientific Reports
DOI 10.1038/s41598-020-66333-xPMC PMC7283481
Open this paper in CoChat — free account

Abstract

Lung cancer is one of the major causes of cancer-related deaths in many countries around the world, and its histopathological diagnosis is crucial for deciding on optimum treatment strategies. Recently, Artificial Intelligence (AI) deep learning models have been widely shown to be useful in various medical fields, particularly image and pathological diagnoses; however, AI models for the pathological diagnosis of pulmonary lesions that have been validated on large-scale test sets are yet to be seen. We trained a Convolution Neural Network (CNN) based on the EfficientNet-B3 architecture, using transfer learning and weakly-supervised learning, to predict carcinoma in Whole Slide Images (WSIs) using a training dataset of 3,554 WSIs. We obtained highly promising results for differentiating between lung carcinoma and non-neoplastic with high Receiver Operator Curve (ROC) area under the curves (AUCs) on four independent test sets (ROC AUCs of 0.975, 0.974, 0.988, and 0.981, respectively). Development and validation of algorithms such as ours are important initial steps in the development of software suites that could be adopted in routine pathological practices and potentially help reduce the burden on pathologists.

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 · 10.1038/s41598-020-66333-x

Cites

9 of 9 references verified0 citation errors · 0 not checkable

Statistics

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

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Kanavati et al. emphasize that deep-learning systems can provide critical support during surgery by offering real-time guidance.

    “Kanavati et al. emphasize how deep-learning (DL) systems can provide critical support during surgery by offering real-time guidance.”

    From Artificial Intelligence in Oncologic Thoracic Surgery: Clinical Decision Support and Emerging Applications (2026) · cites this paper as [47]

    The Introduction says deep learning can aid pathologists in obtaining diagnoses, and the Discussion describes potential uses in assessing lung tissue. Neither section describes real-time guidance during surgery; the paper concerns analysis of histopathology slides, including slides from surgical specimens. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • Trained on large, high-quality datasets, deep-learning models are able to delineate tumor margins intraoperatively with a level of accuracy that exceeds human visual assessment.

    “Trained on large, high-quality datasets, these DL models are able to delineate tumor margins intraoperatively with a level of accuracy that exceeds human visual assessment.”

    From Artificial Intelligence in Oncologic Thoracic Surgery: Clinical Decision Support and Emerging Applications (2026) · cites this paper as [47]

Self-check

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

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

Cites

Flagged first, then by citations

Research works printed but not linked: 28

  1. [2]Chang HY Artificial intelligence in pathology Journal of pathology and translational medicine 2019 53 1 10.4132/jptm.2018.12.16 30599506 PMC6344799 · DOI
  2. [3]Fuchs TJ Buhmann JM Computational pathology: Challenges and promises for tissue analysis Computerized Medical Imaging and Graphics 2011 35 515 530 10.1016/j.compmedimag.2011.02.006 21481567 · DOI
  3. [4]Hou, L. et al. Patch-based convolutional neural network for whole slide tissue image classification. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2424–2433 (2016). PMC5085270 27795661
  4. [7]Kraus OZ Ba JL Frey BJ Classifying and segmenting microscopy images with deep multiple instance learning Bioinformatics 2016 32 i52 i59 10.1093/bioinformatics/btw252 27307644 PMC4908336 · DOI
  5. [9]Luo X Comprehensive computational pathological image analysis predicts lung cancer prognosis Journal of Thoracic Oncology 2017 12 501 509 10.1016/j.jtho.2016.10.017 27826035 PMC5462113 · DOI
  6. [10]Coudray N Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning Nature medicine 2018 24 1559 1567 10.1038/s41591-018-0177-5 PMC9847512 30224757 · DOI · Paper page
  7. [11]Wei JW Pathologist-level classification of histologic patterns on resected lung adenocarcinoma slides with deep neural networks Scientific reports 2019 9 1 8 10.1038/s41598-018-37186-2 30833650 PMC6399447 · DOI · Paper page
  8. [12]Gertych A Convolutional neural networks can accurately distinguish four histologic growth patterns of lung adenocarcinoma in digital slides Scientific reports 2019 9 1483 10.1038/s41598-018-37638-9 30728398 PMC6365499 · DOI
  9. [14]Saltz J Spatial organization and molecular correlation of tumor-infiltrating lymphocytes using deep learning on pathology images Cell reports 2018 23 181 193 10.1016/j.celrep.2018.03.086 29617659 PMC5943714 · DOI · Paper page
  10. [15]Campanella G Clinical-grade computational pathology using weakly supervised deep learning on whole slide images Nature medicine 2019 25 1301 1309 10.1038/s41591-019-0508-1 PMC7418463 31308507 · DOI · Paper page
  11. [16]Iizuka O Deep learning models for histopathological classification of gastric and colonic epithelial tumours Scientific Reports 2020 10 1 11 10.1038/s41598-020-58467-9 32001752 PMC6992793 · DOI
  12. [17]Zhou Z-H A brief introduction to weakly supervised learning National Science Review 2018 5 44 53 10.1093/nsr/nwx106 · DOI · Paper page
  13. [18]Dietterich TG Lathrop RH Lozano-Pérez T Solving the multiple instance problem with axis-parallel rectangles Artificial intelligence 1997 89 31 71 10.1016/S0004-3702(96)00034-3 · DOI · Paper page
  14. [19]Andrews, S., Hofmann, T. & Tsochantaridis, I. Multiple instance learning with generalized support vector machines. Eighteenth national conference on Artificial intelligence 943–944 (2002).
  15. [20]Zhang, C., Platt, J. C. & Viola, P. A. Multiple instance boosting for object detection. Advances in neural information processing systems 1417–1424 (2006).
  16. [21]Babenko B Yang M-H Belongie S Robust object tracking with online multiple instance learning IEEE transactions on pattern analysis and machine intelligence 2010 33 1619 1632 10.1109/TPAMI.2010.226 21173445 · DOI · Paper page
  17. [22]Sudharshan P Multiple instance learning for histopathological breast cancer image classification Expert Systems with Applications 2019 117 103 111 10.1016/j.eswa.2018.09.049 · DOI · Paper page
  18. [24]Grossman RL Toward a shared vision for cancer genomic data New England Journal of Medicine 2016 375 1109 1112 10.1056/NEJMp1607591 27653561 PMC6309165 · DOI · Paper page
  19. [25]Mitsudomi T Gefitinib versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor receptor (wjtog3405): an open label, randomised phase 3 trial The lancet oncology 2010 11 121 128 10.1016/S1470-2045(09)70364-X 20022809 · DOI · Paper page
  20. [26]Hida T Alectinib versus crizotinib in patients with alk-positive non-small-cell lung cancer (j-alex): an open-label, randomised phase 3 trial The Lancet 2017 390 29 39 10.1016/S0140-6736(17)30565-2 28501140 · DOI
  21. [27]Kreyberg L Main histological types of primary epithelial lung tumours British journal of cancer 1961 15 206 10.1038/bjc.1961.25 13754309 PMC2070911 · DOI
  22. [28]Wahbah M Boroumand N Castro C El-Zeky F Eltorky M Changing trends in the distribution of the histologic types of lung cancer: a review of 4,439 cases Annals of diagnostic pathology 2007 11 89 96 10.1016/j.anndiagpath.2006.04.006 17349566 · DOI
  23. [30]National cancer institute clinical proteomic tumor analysis consortium (cptac). radiology data from the clinical proteomic tumor analysis consortium lung squamous cell carcinoma [cptac-lscc] collection [data set], 10.7937/k9/tcia.2018.6emub5l2 (2018). · DOI
  24. [31]Kingma, D. P. & Ba, J. Adam: A method for stochastic optimization. In Bengio, Y. & LeCun, Y. (eds) 3rd International Conference on Learning Representations, Conference Track Proceedings (2015).
  25. [33]Pedregosa F Scikit-learn: Machine learning in Python Journal of Machine Learning Research 2011 12 2825 2830
  26. [35]Efron, B. & Tibshirani, R. J. An introduction to the bootstrap (CRC press, 1994).
  27. [36]Timmermans WMC Van Laar JAM Van Hagen PM Van Zelm MC Immunopathogenesis of granulomas in chronic autoinflammatory diseases Clinical & translational immunology 2016 5 e118 10.1038/cti.2016.75 28090320 PMC5192066 · DOI
  28. [37]Shah KK Pritt BS Alexander MP Histopathologic review of granulomatous inflammation Journal of clinical tuberculosis and other Mycobacterial Diseases 2017 7 1 12 10.1016/j.jctube.2017.02.001 31723695 PMC6850266 · DOI

Cited by

Sorted by citations

Authors
Kanavati, F., Toyokawa, G., Momosaki, S., Rambeau, M., Kozuma, Y., Shoji, F., Yamazaki, K., Takeo, S., Iizuka, O., & Tsuneki, M.
Venue
Scientific Reports
Publisher
Springer Science and Business Media LLC
Published
2020-06-09
Volume
10
Issue
1
ISSN
2045-2322
DOI
10.1038/s41598-020-66333-x
PMC
PMC7283481
Cited by
218
References
9
Language
en
Licence

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