CoChat
Sign Up
Cited by 34Medical Image Segmentation TechniquesCOVID-19 diagnosis using AIMedical Imaging and Analysis

Three-stage segmentation of lung region from CT images using deep neural networks

Open access
Osadebey, Michael, Andersen, Hilde K., Waaler, Dag, Fossaa, Kristian, Martinsen, Anne C. T., Pedersen, Marius·2021·BMC Medical Imaging
DOI 10.1186/s12880-021-00640-1PMC PMC8280386
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Lung region segmentation is an important stage of automated image-based approaches for the diagnosis of respiratory diseases. Manual methods executed by experts are considered the gold standard, but it is time consuming and the accuracy is dependent on radiologists’ experience. Automated methods are relatively fast and reproducible with potential to facilitate physician interpretation of images. However, these benefits are possible only after overcoming several challenges. The traditional methods that are formulated as a three-stage segmentation demonstrate promising results on normal CT data but perform poorly in the presence of pathological features and variations in image quality attributes. The implementation of deep learning methods that can demonstrate superior performance over traditional methods is dependent on the quantity, quality, cost and the time it takes to generate training data. Thus, efficient and clinically relevant automated segmentation method is desired for the diagnosis of respiratory diseases. Methods: We implement each of the three stages of traditional methods using deep learning methods trained on five different configurations of training data with ground truths obtained from the 3D Image Reconstruction for Comparison of Algorithm Database (3DIRCAD) and the Interstitial Lung Diseases (ILD) database. The data was augmented with the Lung Image Database Consortium (LIDC-IDRI) image collection and a realistic phantom. A convolutional neural network (CNN) at the preprocessing stage classifies the input into lung and none lung regions. The processing stage was implemented using a CNN-based U-net while the postprocessing stage utilize another U-net and CNN for contour refinement and filtering out false positives, respectively. Results: The performance of the proposed method was evaluated on 1230 and 1100 CT slices from the 3DIRCAD and ILD databases. We investigate the performance of the proposed method on five configurations of training data and three configurations of the segmentation system; three-stage segmentation and three-stage segmentation without a CNN classifier and contrast enhancement, respectively. The Dice-score recorded by the proposed method range from 0.76 to 0.95. Conclusion: The clinical relevance and segmentation accuracy of deep learning models can improve though deep learning-based three-stage segmentation, image quality evaluation and enhancement as well as augmenting the training data with large volume of cheap and quality training data. We propose a new and novel deep learning-based method of contour refinement.

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 · 10.1186/s12880-021-00640-1

Cites

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?

  • The model lacks spatial reasoning to tell air cavities apart from lung tissue.

    “the model lacks spatial reasoning to tell them apart. 29”

    From Uncertainty quantification of U‐Net based segmentation tool using conformal prediction (2026) · cites this paper as [29]

    The Introduction (page unavailable) says that bowel gas and subcutaneous cavities can have grayscale values similar to the lung region, making automated segmentation challenging. It does not say the model lacks spatial reasoning or is unable to distinguish those structures from lung tissue; the Discussion’s “Reduction of false positives” section instead reports that its CNN classifiers reduce false positives. (LLM, conf=0.83)

    Experimental — accuracy not yet measured

Self-check

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

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

The citation lists haven’t been fetched yet

References 46 · Cited by 34

Research works printed but not linked: 39

  1. [1]Hu S Hoffman EA Reinhardt JM Automatic lung segmentation for accurate quantitation of volumetric x-ray ct images IEEE Trans Med Imaging 2001 20 6 490 498 10.1109/42.929615 11437109 · DOI
  2. [2]Pakdemirli E, Mandalia U, Monib S. Positive chest ct features in patients with covid-19 pneumonia and negative real-time polymerase chain reaction test. Cureus. 2020;12(8). PMC7444987 32850265
  3. [3]Liu C Zhao R Xie W Pang M Pathological lung segmentation based on random forest combined with deep model and multi-scale superpixels Neural Process Lett 2020 52 2 1631 1649 10.1007/s11063-020-10330-8 PMC7413019 32837245 · DOI
  4. [4]Mittal A Hooda R Sofat S Lung field segmentation in chest radiographs: a historical review, current status, and expectations from deep learning IET Image Proc 2017 11 11 937 952 10.1049/iet-ipr.2016.0526 · DOI
  5. [5]Kamble B, Sahu SP, Doriya R. A review on lung and nodule segmentation techniques. In: Advances in Data and Information Sciences, pp. 555–565. Springer, New York. 2020.
  6. [7]Mansoor A Bagci U Xu Z Foster B Olivier KN Elinoff JM Suffredini AF Udupa JK Mollura DJ A generic approach to pathological lung segmentation IEEE Trans Med Imaging 2014 33 12 2293 2310 10.1109/TMI.2014.2337057 25020069 PMC5542015 · DOI
  7. [8]Shi Z, Ma J, Zhao M, Liu Y, Feng Y, Zhang M, He L, Suzuki K. Many is better than one: an integration of multiple simple strategies for accurate lung segmentation in ct images. BioMed Res Int. 2016. PMC5011243 27635395
  8. [9]Sun L, Peng Z, Wang Z, Pu H, Guo L, Yuan G, Yin F, Pu T. Automatic lung segmentation in chest ct image using morphology. In: 9th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Sensing and Imaging, 2019;vol. 10843, p. 108431. International Society for Optics and Photonics.
  9. [10]Javaid M Javid M Rehman MZU Shah SIA A novel approach to cad system for the detection of lung nodules in ct images Comput Methods Prog Biomed 2016 135 125 139 10.1016/j.cmpb.2016.07.031 27586486 · DOI
  10. [11]Nithila EE Kumar S Segmentation of lung from ct using various active contour models Biomed Signal Process Control 2019 47 57 62 10.1016/j.bspc.2018.08.008 · DOI
  11. [12]Chan TF Vese LA Active contours without edges IEEE Trans Image Process 2001 10 2 266 277 10.1109/83.902291 18249617 · DOI
  12. [13]Caselles V Kimmel R Sapiro G Geodesic active contours Int J Comput Vision 1997 22 1 61 79 10.1023/A:1007979827043 · DOI
  13. [14]Otsu N A threshold selection method from gray-level histograms IEEE Trans Syst Man Cybern 1979 9 1 62 66 10.1109/TSMC.1979.4310076 · DOI
  14. [15]Yim Y Hong H Correction of segmented lung boundary for inclusion of pleural nodules and pulmonary vessels in chest ct images Comput Biol Med 2008 38 8 845 857 10.1016/j.compbiomed.2008.04.012 18589408 · DOI
  15. [16]Liu C Pang M Automatic lung segmentation based on image decomposition and wavelet transform Biomed Signal Process Control 2020 61 102032 10.1016/j.bspc.2020.102032 · DOI
  16. [17]Shojaii R, Alirezaie J, Babyn P. Automatic lung segmentation in ct images using watershed transform. In: IEEE International Conference on Image Processing 2005, 2005;vol. 2, p. 1270. IEEE.
  17. [20]Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015;pp. 3431–3440. 27244717
  18. [21]Skourt BA El Hassani A Majda A Lung ct image segmentation using deep neural networks Procedia Comput Sci 2018 127 109 113 10.1016/j.procs.2018.01.104 · DOI
  19. [22]Pang T, Guo S, Zhang X, Zhao L. Automatic lung segmentation based on texture and deep features of hrct images with interstitial lung disease. BioMed Research International. 2019. PMC6907046 31871932
  20. [23]Ghosh S, Sil S, Gomes RM, Dey M. Using convolutions and image processing techniques to segment lungs from ct data. In: Emerging Technology in Modelling and Graphics, pp. 129–136. Springer, New York 2020.
  21. [24]Armato SRD Mclennan G Bidaut L Mcnitt-gray MF Meyer CR Reeves AP The lung image database consortium (lidc) and image database resource initiative (idri): A completed reference database of lung nodules on ct scans Med Phys 2011 38 915 31 10.1118/1.3528204 21452728 PMC3041807 · DOI · Paper page
  22. [27]Afadzi M Lysvik EK Andersen HK Martinsen ACT Ultra-low dose chest computed tomography: effect of iterative reconstruction levels on image quality Eur J Radiol 2019 114 62 68 10.1016/j.ejrad.2019.02.021 31005179 · DOI
  23. [28]Xu M Qi S Yue Y Teng Y Xu L Yao Y Qian W Segmentation of lung parenchyma in ct images using cnn trained with the clustering algorithm generated dataset Biomed Eng Online 2019 18 1 2 10.1186/s12938-018-0619-9 30602393 PMC6317251 · DOI
  24. [29]Osadebey M, Pedersen M, Waaler D. Learning-based segmentation of optic disc in retinal images using clustering trees and local mode filtering. In: 14th Multi Conference on Computer Science and Information Systems 2020: IADIS International Conferences Interfaces and Human Computer Interaction 2020; Game and Entertainment Technologies 2020; and Computer Graphics, Visualization, Computer Vision and Image Processing 2020;pp. 124–130. IADIS.
  25. [30]Zappia L Oshlack A Clustering trees: a visualization for evaluating clusterings at multiple resolutions GigaScience 2018 7 7 083 10.1093/gigascience/giy083 PMC6057528 30010766 · DOI · Paper page
  26. [31]Osadebey M Pedersen M Arnold D Wendel-Mitoraj K Local indicators of spatial autocorrelation (lisa): application to blind noise-based perceptual quality metric index for magnetic resonance images J Imaging 2019 5 1 20 10.3390/jimaging5010020 PMC8320873 34465703 · DOI
  27. [32]Al-Ameen Z Sulong G A new algorithm for improving the low contrast of computed tomography images using tuned brightness controlled single-scale retinex Scanning 2015 37 2 116 125 10.1002/sca.21187 25663630 · DOI
  28. [33]Hofmanninger J Prayer F Pan J Röhrich S Prosch H Langs G Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem Eur Radiol Exp 2020 4 1 1 13 10.1186/s41747-020-00173-2 32814998 PMC7438418 · DOI
  29. [34]Pakdemirli E New subspecialty or close ally emerging on the radiology horizon: artificial intelligence Acta Radiologica Open 2019 8 3 2058460119837498 10.1177/2058460119837498 30923633 PMC6431767 · DOI
  30. [35]Hesamian MH Jia W He X Kennedy P Deep learning techniques for medical image segmentation: achievements and challenges J Digit Imaging 2019 32 4 582 596 10.1007/s10278-019-00227-x 31144149 PMC6646484 · DOI
  31. [36]Lemley J Bazrafkan S Corcoran P Deep learning for consumer devices and services: pushing the limits for machine learning, artificial intelligence, and computer vision IEEE Consum Electron Mag 2017 6 2 48 56 10.1109/MCE.2016.2640698 · DOI
  32. [37]You C, Yang L, Zhang Y, Wang G. Low-dose ct via deep cnn with skip connection and network-in-network. In: Developments in X-Ray Tomography XII, 2019;vol. 11113, p. 111131. International Society for Optics and Photonics
  33. [38]Wu, D, Kim K, Fakhri GE, Li Q. A cascaded convolutional neural network for X-ray low-dose ct image denoising. arXiv preprint arXiv:1705.04267, 2017. · arXiv
  34. [39]Bai M, Urtasun R. Deep watershed transform for instance segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017;pp. 5221–5229.
  35. [40]Liu P Zhang H Lian W Zuo W Multi-level wavelet convolutional neural networks IEEE Access 2019 7 74973 74985 10.1109/ACCESS.2019.2921451 · DOI
  36. [41]Pakdemirli E, Mandalia U, Monib S. Characteristics of chest ct images in patients with Covid-19 pneumonia in London, UK. Cureus. 2020;12(9). PMC7482987 32923301
  37. [42]Pakdemirli E Perception of artificial intelligence (ai) among radiologists Acta radiologica open 2019 8 9 2058460119878662 10.1177/2058460119878662 31632696 PMC6769222 · DOI
  38. [43]Smith SM Brady JM Susan-a new approach to low level image processing Int J Comput Vision 1997 23 1 45 78 10.1023/A:1007963824710 · DOI · Paper page
  39. [46]Oulefki A Agaian S Trongtirakul T Laouar AK Automatic covid-19 lung infected region segmentation and measurement using ct-scans images Pattern Recogn 2021 114 107747 10.1016/j.patcog.2020.107747 PMC7605758 33162612 · DOI

Authors
Osadebey, M., Andersen, H. K., Waaler, D., Fossaa, K., Martinsen, A. C. T., & Pedersen, M.
Affiliations
  • Norwegian University of Science and Technology (NO)
  • Oslo University Hospital (NO)
  • OsloMet – Oslo Metropolitan University (NO)
Corresponding author
Osadebey, Michael
ORCID
Venue
BMC Medical Imaging
Publisher
Springer Science and Business Media LLC
Published
2021-07-15
Volume
21
Issue
1
Pages
112-112
ISSN
1471-2342 (ISSN-L)
DOI
10.1186/s12880-021-00640-1
PMC
PMC8280386
OpenAlex
W3181291819
Cited by
34
References
46
Citations by year
2026: 7 · 2025: 8 · 2024: 5 · 2023: 12 · 2022: 1 · 2021: 1
FWCI
1.71
Type
journal article
Language
en
Open access
Licence
Locations
  • BMC Medical Imaging (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • NORA - Norwegian Open Research Archives (submittedVersion, other-oa)
  • NORA - Norwegian Open Research Archives (submittedVersion, cc-by)
Funders
European Research Consortium for Informatics and Mathematics
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
  • Physical Sciences › Engineering › Biomedical Engineering › Medical Imaging and Analysis
Keywords
lung segmentation, CT images, deep learning, convolutional neural networks, U-Net, contour refinement, data augmentation, interstitial lung disease, respiratory disease diagnosis, LIDC-IDRI, image enhancement
MeSH terms
Deep Learning, Algorithms, Humans, Lung, Lung Diseases, Tomography, X-Ray Computed, 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