Skip to main content

Advertisement

SpringerLink
Log in

Characterization of Architectural Distortion in Mammograms Based on Texture Analysis Using Support Vector Machine Classifier with Clinical Evaluation

  • Published:
Journal of Digital Imaging Aims and scope Submit manuscript

Abstract

Architecture distortion (AD) is an important and early sign of breast cancer, but due to its subtlety, it is often missed on the screening mammograms. The objective of this study is to create a quantitative approach for texture classification of AD based on various texture models, using support vector machine (SVM) classifier. The texture analysis has been done on the region of interest (ROI) selected from the original mammogram. A comprehensive analysis has been done on samples from three databases; out of which, two data sets are from the public domain, and the third data set is for clinical evaluation. The public domain databases are IRMA version of digital database for screening mammogram (DDSM) and Mammographic Image Analysis Society (MIAS). For clinical evaluation, the actual patient’s database has been obtained from ACE Healthways, Diagnostic Centre Ludhiana, India. The significant finding of proposed study lies in appropriate selection of the size of ROIs. The experiments have been done on fixed size of ROIs as well as on the ground truth (variable size) ROIs. Best results pertain to an accuracy of 92.94 % obtained in case of DDSM database for fixed-size ROIs. In case of MIAS database, an accuracy of 95.34 % is achieved in AD versus non-AD (normal) cases for ground truth ROIs. Clinically, an accuracy of 88 % was achieved for ACE dataset. The results obtained in the present study are encouraging, as optimal result has been achieved for the proposed study in comparison with other related work in the same area.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

References

  1. Breast Cancer in India: Available at http://global.umich.edu/2013/02/the-growing-problem-of-breast-cancer-in-india. Accessed 13 January, 2014

  2. Population based cancer registry: Available at http://www.pbcrindia.org/. Accessed 13 January, 2014

  3. American College of Radiology: The ACR breast imaging reporting and data system 2003. Available at http://www.acr.orgdepartments/stand_accred/birads/contents.html. Accessed 25 January, 2014

  4. Rao AR, Jain CR: Computerized flow field analysis: oriented texture fields. IEEE Trans. Pattern Anal. Mach. Intell. 14:693–709, 1992

    Article  Google Scholar 

  5. Guo Q, Shao J, Ruiz V: Investigation of support vector machine for the detection of architectural distortion in mammographic images. Journal of Physics: Conference Series 15:88–94, 2005

    Google Scholar 

  6. Nandi RJ, Nandi AK, Rangayyan RM, Scutt D: Classification of breast masses in mammograms using genetic programming and feature selection. Med Biol Eng Comput 44:683–694, 2006

    Article  CAS  PubMed  Google Scholar 

  7. Ayres FJ, Rangayyan RM: Gabor filters and phase portrait for the detection of architectural distortion in mammograms. Med Biol Eng Comput 44:883–894, 2006

    Article  PubMed  Google Scholar 

  8. Banik S, Rangayyan RM, Desautels JEL: Computer Aided Detection of Architectural Distortion in Prior Mammograms of Interval Cancer. J Digit Imaging 23:611–631, 2010

    Article  PubMed Central  PubMed  Google Scholar 

  9. Buciu L, Gacsadi A: Directional features for automatic tumor classification of mammogram images. Biomed Signal Process Control 6:370–378, 2001

    Article  Google Scholar 

  10. Eltoukhy MM, Faye I, Samir BB: A statistical based feature extraction method for breast cancer diagnosis in digital mammogram using multi-resolution representation. Comput Biol Med 42:123–128, 2012

    Article  Google Scholar 

  11. Harlick RM, Shanmugan K, Dinstein IH: Texture Features for image classification. IEEE Trans. Syst Man Cybern. Syst. 3:610–621, 1973

    Article  Google Scholar 

  12. Weszka JS, Dyer CR, Rosenfeild A: A Comparative Study of Texture Measures for Terrain Classification. IEEE Trans. Syst Man Cybern. Syst. 6:269–285, 1976

    Article  Google Scholar 

  13. Mandelbrot BB: The Fractal Geometry of Nature. W.H. Freeman and Company, New York, 1986

    Google Scholar 

  14. Tourassi GD, Delong DM, Floyd CE: A study on the computerized fractal analysis of architectural distortion in screening mammograms. Phys. Med Biol 51:1299–1312, 2006

    Article  PubMed  Google Scholar 

  15. Fukunaga K: Introduction to Statistical Pattern Recognition, 2nd edition. Academic, San Diego, 1997

    Google Scholar 

  16. Mathworks.com: Available at http://www.mathworks.com/help/toolbox/Stats/70-71. Accessed 10 Feb. 2014

  17. Cortes C, Vapnik V: Support vector networks. Mach Learn 20:273–297, 1995

    Google Scholar 

  18. Oliveira JED, Guld M, Araujo A, Ott B, Deserno TM: Towards a standard reference database for computer-aided mammography, Proceedings of SPIE Medical Imaging 69151Y, 2008

  19. Suckling J, Parker J, Dance J, Astley DR, Hutt S, Boggis I, Ricketts C, Stamatakis I, Cerneaz E, Kok N, Taylor SN, Betal P, Savage D: The Mammographic Image Analysis Society digital mammogram database. Proceedings of 2nd International Workshop on Digital Mammography, New York, USA, 1994, pp 375–378

    Google Scholar 

  20. Kamra A, Jain VK, Singh S: Extraction of orientation field using Gabor Filter and Gradient based approach for the detection of subtle signs in mammograms. J. Med. Imaging & Health Infor. 4:374–381, 2014

    Article  Google Scholar 

  21. Li H, Giger ML, Huo Z, Olufunmilayo I, Li O, Barbara L, Weber L, Bonta L: Computerized analysis of mammographic parenchyma patterns for assessing breast cancer risk: Effect of ROI size and location. J. Med. Phys. 31:549–555, 2004

    Article  Google Scholar 

  22. Robert C, Ike III, Singh S, Harrawood B, Tourassi GD: Effect of ROI Size on the Performance of an Information-Theoretic CAD System in Mammography: Multi-size Fusion Analysis. Proceedings of SPIE, Bellingham, USA, 2008, p 6915

    Google Scholar 

  23. Bovis K, Singh S: Detection of masses in mammograms using texture features. Proceedings of the 15th international conference on pattern recognition, Barcelona, Spain, 2000, pp 267–270

    Google Scholar 

  24. Rangayyan RM, Nguyen TM: Fractal Analysis of contours of breast masses in mammograms. J Digit Imaging 20:223–237, 2007

    Article  PubMed Central  PubMed  Google Scholar 

  25. Serrano RC, Conci A, Zamith M, Lima R: About the feasibility of Hurst coefficient in thermal images for early diagnosis of breast diseases. Proceedings of the 11th Pan-American Congress of Applied Mechanics, Parana, Brazil, 2010

    Google Scholar 

  26. Theodoridis S, Koutroumbas K: An Introduction to Pattern Recognition: A MATLAB Approach, 4th edition. Academic, Burlington, 2010

    Google Scholar 

  27. Sahiner B, Chan HP, Petrick N, Wagner RF, Hadjiiski L: Feature Selection and classifier performance in computer aided diagnosis: the effect of finite sample size. J. Med. Phys. 27:1509–1522, 2000

    Article  CAS  Google Scholar 

  28. Yang X, Cao A, Song Q, Schaefer G, Su Y: Vicinal support vector classifier using supervised kernel based clustering. Artif Intell Med 60:189–196, 2014

    Article  CAS  PubMed  Google Scholar 

  29. Metz CE: Evaluation of digital mammography by ROC analysis. Proceedings of the 3rd International Workshop on Digital Mammography, Chicago, 1996, pp 61–68

    Google Scholar 

  30. Witten IH, Frank E: Data Mining: practical machine learning tools and technique, 2nd edition. Morgan Kauffman, Burlington, 2005

    Google Scholar 

  31. Cross Validation: Available at http://research.cs.tamu.edu/prism/lectures/iss/iss_l13.pdf. Accessed 25 March, 2015

  32. Jasionowska M, Przedaskowski A, Rutczynska A, Wroblewska A: A two step method for detection of architectural distortion in mammograms. Information Technology in Biomedicine. Advances in Intelligent and Soft Computing. 69:73–84, 2010

    Article  Google Scholar 

  33. Minavathi, Murali M, Dinesh MS: Model based approach for Detection of Architectural Distortions and Spiculated Masses in Mammograms. Int. J Comput. Science Engg 3:3534–3546, 2011

    Google Scholar 

Download references

Acknowledgments

The authors are highly thankful to Dr Ravinder Sidhu, radiologist at ACE Healthways, Ludhiana, India, and Dr Ramandeep Singh, radiologist at Delta Imaging Centre, Ludhiana, India for their support in this work. The authors also thank Dr. Thomas Deserno, Department of Medical Informatics, Aachen University of Technology, Aachen, North Rhine-Westphalia, Germany, for providing the image retrieval in medical application (IRMA) version of DDSM database. The author also thank ACE Healthways, Diagnostic Centre, Ludhiana (India), for providing the actual patient data for carrying out the clinical evaluation of this study. The authors also feel gratitude towards anonymous reviewers for providing substantial and useful review, which led to important improvement in the present manuscript

Conflict of Interest

The authors declare that they have no competing interests.

Author information

Authors and Affiliations

  1. Department of Information Technology, Guru Nanak Dev Engineering College, Ludhiana, India

    Amit Kamra

  2. Department of Electrical and Instrumentation Engineering, SLIET, Deemed University, Sangrur, India

    V K Jain

  3. Department of Computer Science and Engineering, UIET, Panjab University, Chandigarh, India

    Sukhwinder Singh

  4. Department of Radiology, ACE Healthways Diagnostics Centre, Ludhiana, India

    Sunil Mittal

Authors
  1. Amit Kamra
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. V K Jain
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sukhwinder Singh
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sunil Mittal
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Sukhwinder Singh.

Additional information

Highlights

1. Architectural distortion is the most difficulty abnormality to classify due to its varying attributes.

2. The system is trained and tested with images from three databases. Two databases viz. MIAS and DDSM are standard public domain databases. For clinical evaluation, the experiment has been done on actual patient database obtained from ACE Healthways, Diagnostic Centre Ludhiana, India.

3. The system has been tested with two types of ROI viz. fixed size and Ground (varying) truth.

4. Three features models have been used, GLCM, Fractal features and Fourier power spectrum.

5. A unique combination of texture based features has been proposed after reducing with Stepwise Regression based Feature Selection method.

6. The results have been validated with different performance evaluation parameters using support vector machine classifier using sequential minimal optimization.

7. The results have been validated statistically based on various statistical parameters.

8. Above all the accuracy of proposed classification system is quite high and comparable to work of other researchers

9. Results of clinical evaluation done on actual patient care signify that proposed study can be very useful in providing second opinion to the radiologists.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kamra, A., Jain, V.K., Singh, S. et al. Characterization of Architectural Distortion in Mammograms Based on Texture Analysis Using Support Vector Machine Classifier with Clinical Evaluation. J Digit Imaging 29, 104–114 (2016). https://doi.org/10.1007/s10278-015-9807-3

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10278-015-9807-3

Keywords

Search

Navigation