Advertisement

Medical & Biological Engineering & Computing

, Volume 56, Issue 8, pp 1447–1458 | Cite as

Angular relational signature-based chest radiograph image view classification

  • K. C. Santosh
  • Laurent Wendling
Original Article

Abstract

In a computer-aided diagnosis (CAD) system, especially for chest radiograph or chest X-ray (CXR) screening, CXR image view information is required. Automatically separating CXR image view, frontal and lateral can ease subsequent CXR screening process, since the techniques may not equally work for both views. We present a novel technique to classify frontal and lateral CXR images, where we introduce angular relational signature through force histogram to extract features and apply three different state-of-the-art classifiers: multi-layer perceptron, random forest, and support vector machine to make a decision. We validated our fully automatic technique on a set of 8100 images hosted by the U.S. National Library of Medicine (NLM), National Institutes of Health (NIH), and achieved an accuracy close to 100%. Our method outperforms the state-of-the-art methods in terms of processing time (less than or close to 2 s for the whole test data) while the accuracies can be compared, and therefore, it justifies its practicality.

Graphical Abstract

Interpreting chest X-ray (CXR) through the angular relational signature.

Keywords

Automation Computer-aided diagnosis system Medical imaging Chest radiograph Angular relational signature Force histogram Frontal view Lateral view Classification Multi-layer perception Random forest Support vector machine 

Notes

Acknowledgments

Authors would like to give thanks to Dr. Sameer Antani, staff scientist, Lister Hill National Center for Biomedical Communications (LHNCBC) for his advice and the dataset owned by the U.S. National Library of Medicine, National Institutes of Health (NIH), MD.

Compliance with ethical standards

This article does not contain any studies with human participants or animals performed by any of the authors.

Conflict of interest

Authors declare that they have no conflicts of interest.

References

  1. 1.
    (2006) Improving the diagnosis and treatment of smear-negative pulmonary and extrapulmonary tuberculosis among adults and adolescents: recommendations for HIV-prevalent and resource-constrained settings. World Health Organization GenevaGoogle Scholar
  2. 2.
    Arimura H, Katsuragawa S, Li Q, Ishida T, Doi K (2002) Development of a computerized method for identifying the posteroanterior and lateral views of chest radiographs by use of a template matching technique. Med Phys 29(7):1556–1561.  https://doi.org/10.1118/1.1487426 CrossRefPubMedGoogle Scholar
  3. 3.
    Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press, inc., New YorkGoogle Scholar
  4. 4.
    Boone JM, Seshagiri S, Steiner RM (1992) Recognition of chest radiograph orientation for picture archiving and communications systems display using neural networks. J Digit Imaging 5(3):190–193.  https://doi.org/10.1007/BF03167769 CrossRefPubMedGoogle Scholar
  5. 5.
    Bosch A, Zisserman A, Munoz X (2007) Representing shape with a spatial pyramid kernel. In: ACM international conference on image and video retrieval, pp 401–408Google Scholar
  6. 6.
    Breiman L (2001) Random forests. Mach Learn 45(1):5–32CrossRefGoogle Scholar
  7. 7.
    Dubois D, Jaulent M (1987) A general approach to parameter evaluation in fuzzy digital pictures. Pattern Recogn Lett 6:251–259CrossRefGoogle Scholar
  8. 8.
    Kao EF, Lee C, Jaw TS, Hsu JS, Liu GC (2006) Projection profile analysis for identifying different views of chest radiographs. Acad Radiol 13(4):518–525.  https://doi.org/10.1016/j.acra.2006.01.009 CrossRefPubMedGoogle Scholar
  9. 9.
    Karargyris A, Siegelman J, Tzortzis D, Jaeger S, Candemir S, Xue Z, Santosh KC, Vajda S, Antani SK, Folio L, Thoma GR (2016) Combination of texture and shape features to detect pulmonary abnormalities in digital chest X-rays. Int J Comput Assist Radiol Surg 11(1):99–106.  https://doi.org/10.1007/s11548-015-1242-x CrossRefPubMedGoogle Scholar
  10. 10.
    Keerthi S, Shevade S, Bhattacharyya C, Murthy K (2001) Improvements to Platt’s SMO algorithm for SVM classifier design. Neural Comput 13(3):637–649CrossRefGoogle Scholar
  11. 11.
    Krishnapuram R, Keller JM, Ma Y (1993) Quantitative analysis of properties and spatial relations of fuzzy image regions. IEEE Trans Fuzzy Syst 1(3):222–233.  https://doi.org/10.1109/91.236554 CrossRefGoogle Scholar
  12. 12.
    Lehmann TM, Güld MO, Keysers D, Schubert H, Kohnen M, Wein BB (2003) Determining the view of chest radiographs. J Digit Imaging 16(3):280–291.  https://doi.org/10.1007/s10278-003-1655-x CrossRefPubMedPubMedCentralGoogle Scholar
  13. 13.
    Matsakis P (1998) Structural spatial relations and image understanding. Ph.D. thesis, Universite de Toulouse III.  https://doi.org/10.1007/978-3-642-14755-5
  14. 14.
    Matsakis P, Wendling L (1999) A new way to represent the relative position between areal objects. IEEE Trans Pattern Anal Mach Intell 21(7):634–643CrossRefGoogle Scholar
  15. 15.
    Matsakis P, Wendling L, Ni J (2010) A general approach to the fuzzy modeling of spatial relationships. In: Methods for handling imperfect spatial information.  https://doi.org/10.1007/978-3-642-14755-5_3, pp 49–74
  16. 16.
    Pietka E, Huang HK (1992) Orientation correction for chest images. J Digit Imaging 5(3):185–189.  https://doi.org/10.1007/BF03167768 CrossRefPubMedGoogle Scholar
  17. 17.
    Platt J (1998) Fast training of support vector machines using sequential minimal optimization. In: Schoelkopf B, Burges C, Smola A (eds) Advances in kernel methods - support vector learning, MIT Press. http://research.microsoft.com/~jplatt/smo.html
  18. 18.
    Platt JC (1999) Advances in kernel methods. chap. fast training of support vector machines using sequential minimal optimization. MIT Press, Cambridge, pp 185–208. http://dl.acm.org/citation.cfm?id=299094.299105 Google Scholar
  19. 19.
    Santosh KC, Candemir S, Jäger S, Karargyris A, Antani SK, Thoma GR, Folio L (2015) Automatically detecting rotation in chest radiographs using principal rib-orientation measure for quality control. Int J Pattern Recognit Artif Intell 29(2).  https://doi.org/10.1142/S0218001415570013
  20. 20.
    Santosh KC, Vajda S, Antani S, Thoma GR (2016) Edge map analysis in chest X-rays for automatic pulmonary abnormality screening. Int J CARS :1–10.  https://doi.org/10.1007/s11548-016-1359-6
  21. 21.
    Santosh KC, Wendling L (2017) Automated chest X-ray image view classification using force histogram. Springer, Singapore, pp 333–342.  https://doi.org/10.1007/978-981-10-4859-3_30 Google Scholar
  22. 22.
    Schaefer-Prokop C, Neitzel U, Venema H, Uffmann M, Prokop M (2008) Digital chest radiography: an update on modern technology, dose containment and control of image quality. Eur Radiol 18(9):1818–1830CrossRefPubMedPubMedCentralGoogle Scholar
  23. 23.
    Wendling L, Tabbone S, Matsakis P (2002) Fast and robust recognition of orbit and sinus drawings using histograms of forces. Pattern Recogn Lett 23(14):1687–1693CrossRefGoogle Scholar
  24. 24.
    Xue Z, You D, Candemir S, Jaeger S, Antani SK, Long LR, Thoma GR (2015) Chest x-ray image view classification. In: 28Th IEEE international symposium on computer-based medical systems.  https://doi.org/10.1109/CBMS.2015.49, pp 66–71
  25. 25.
    You D, Antani SK, Demner-fushman D, Thoma GR (2014) A contour-based shape descriptor for biomedical image classification and retrieval. In: Coüasnon B, Ringger EK (eds) Document recognition and retrieval XXI, San Francisco, February 5-6, 2014. SPIE Proceedings, vol 9021, pp 90210L–90210L–12. SPIE.  https://doi.org/10.1117/12.2042526

Copyright information

© International Federation for Medical and Biological Engineering 2018

Authors and Affiliations

  1. 1.Department of Computer ScienceThe University of South DakotaVermillionUSA
  2. 2.LIPADEUniversité Paris Descartes (Paris V)Paris Cedex 06France

Personalised recommendations