Advertisement

Research on Biomedical Engineering

, Volume 35, Issue 3–4, pp 241–249 | Cite as

Automatic segmentation of blood vessels in retinal images using 2D Gabor wavelet and sub-image thresholding resulting from image partition

  • Luciana da Silva AmorimEmail author
  • Flávia Magalhães Freitas Ferreira
  • Juliana Reis Guimarães
  • Zélia Myriam Assis Peixoto
Original Article
  • 4 Downloads

Abstract

Purpose

The retina features the only blood vessel network in humans that is visible in a non-invasive imaging method. This, along with uniqueness and stability throughout life in healthy subjects, makes it an ideal target for personal identification methods in biometric systems and also for the screening and diagnosis of diseases. However, retinal images usually present low contrast of the vessels in relation to the retinal background and high level of noise stemming mainly from the acquisition process. This work aims to reduce noise and improve contrast to increase the accuracy of retinal vessel segmentation.

Methods

2D Gabor wavelet (GW) is usually employed to reduce noise and improve vessel contrast in relation to the background. In this work, it is proposed that, before the thresholding, the GW output images are partitioned into 20 sub-images in such a way that each can be treated independently.

Results

The images used were obtained from two public databases, DRIVE and STARE, and the algorithm was developed in MatLab® environment. The proposed approach reached an accuracy of 96.15%, sensitivity of 73.42%, and specificity of 98.30% in DRIVE. In STARE, the accuracy was 94.87%, sensitivity 71.74%, and specificity 96.93%.

Conclusion

The methods proposed by the authors indicate gains in accuracy and specificity in the automatic detection of retinal vessels, in both databases used, when compared with those in the main published works. The accuracy is also higher than the 94.73% in interobserver accuracy previously determined as the gold standard.

Keywords

Low contrast Denoising 2D Gabor wavelet Image partitioning Image thresholding Segmentation of retinal vessels 

Notes

Acknowledgments

The authors would like to thank JJ Staal, AD Hoover, and their colleagues for making their databases publicly available and the Laboratory of Applied Research in Neuroscience of Vision (LAPAN) along with Dr. Ricardo Guimarães Eye Hospital for the technical support.

Funding information

This study was financed in part by the CAPES - Brazil - Finance Code 001.

References

  1. Abramoff MD, Garvin MK, Sonka M. Retinal imaging and image analysis. IEEE Rev Biomed Eng. 2010;3:169–208.  https://doi.org/10.1109/RBME.2010.2084567.CrossRefGoogle Scholar
  2. Akram MU, Tariq A, Nasir S, Khan AS. Gabor wavelet based vessel segmentation in retinal images. In: IEEE symposium on computational intelligence for image processing; 2009 Mar 30 - Apr 2; Nashville, United States of America. USA: IEEE; 2009. p. 116–9.  https://doi.org/10.1109/CIIP.2009.4937890.CrossRefGoogle Scholar
  3. Akram MU, Tariq A, Khan AS. Retinal recognition: personal identification using blood vessels. In: International conference for internet technology and secured transactions; 2011 Dec 11–14; Abu Dhabi, United Arab Emirates. USA: IEEE; 2012; 2011. p. 180–4.Google Scholar
  4. Ali A, Hussain A, Wan Zaki WMD. Vessel extraction in retinal images using automatic thresholding and Gabor wavelet. In: 39th Annual international conference of the IEEE engineering in medicine and biology society (EMBC); 2017 Jul 11–15; Seogwipo, South Korea. USA: IEEE; 2017. p. 365–8.  https://doi.org/10.1109/EMBC.2017.8036838.CrossRefGoogle Scholar
  5. AlZaid E, Shalash WM, Abulkhair MF. Retinal blood vessels segmentation using Gabor filters. In: 2018 1st international conference on computer applications information security (ICCAIS); 2018 Apr 4–6; Riyadh, Saudi Arabia. USA: IEEE; 2018. p. 1–6.  https://doi.org/10.1109/CAIS.2018.8441937.
  6. Arthur AM, Arthur R, Silva AG, Fouto MS, Iano Y, Faria J. Algorithm for predicting macular dysfunction based on moment invariants classification of the foveal avascular zone in functional retinal images. Rev Bras Eng Biomed. 2017;33(4):344–51.  https://doi.org/10.1590/2446-4740.01417.CrossRefGoogle Scholar
  7. Bařina D (2011) Gabor wavelets in image processing. In: Proceedings of the 17th Conference STUDENT EEICT; Apr; Brno, Czech Republic. p. 522–526. arXiv:1602.03308v1.Google Scholar
  8. Fatima J, Syed AM, Akram U. A secure personal identification system based on human retina. In: 2013 IEEE symposium on industrial electronics applications; 2013 Sep 22–25; Kuching, Malaysia. USA: IEEE; 2013. p. 90–5.  https://doi.org/10.1109/ISIEA.2013.6738974.CrossRefGoogle Scholar
  9. Fraz MM, Rudnicka AR, Owen CG, Strachan DP, Barman SA. Automated arteriole and venule recognition in retinal images using ensemble classification. In: 2014 international conference on computer vision theory and applications (VISAPP). Portugal. USA: IEEE; 2015; 2014 Jan 05–08; Lisbon. p. 194–202.  https://doi.org/10.5220/0004733701940202.CrossRefGoogle Scholar
  10. Gou D, Ma T, Wei Y. A novel retinal vessel extraction method based on dynamic scales allocation. In: 2017 2nd international conference on image, vision and computing (ICIVC). Chengdu, China. USA: IEEE; 2017 Jun 2–4. p. 145–9.  https://doi.org/10.1109/ICIVC.2017.7984535.CrossRefGoogle Scholar
  11. Haleem MS, Han L, Jv H, Li B. Fleming a. retinal area detector from scanning laser ophthalmoscope (SLO) images for diagnosing retinal diseases. IEEE Journal of Biomedical and Health Informatics. 2015;19(4):1472–82.  https://doi.org/10.1109/JBHI.2014.2352271.CrossRefGoogle Scholar
  12. Hoover AD, Kouznetsova V, Goldbaum M. Structured analysis of the retina [internet]. California, United States of America; 2000. [cited 2018 Jul 20]. Available from: http://cecas.clemson.edu/~ahoover/stare/.
  13. Kim CH, Aggarwal R. Wavelet transforms in power systems II examples of application to actual power system transients. Power Engineering Journal. 2001;15(4):193–202.  https://doi.org/10.1049/pe:20010404.CrossRefGoogle Scholar
  14. Lienert RT. Inter-observer comparisons of ophthalmoscopic assessment of diabetic retinopathy. Aust N Z J Ophthalmol. 1989;17(4):363–8.  https://doi.org/10.1111/j.1442-9071.1989.tb00555.x.CrossRefGoogle Scholar
  15. Miri MS, Mahloojifar A. A comparison study to evaluate retinal image enhancement techniques. In: 2009 IEEE international conference on signal and image processing applications; 2009 Nov 18–19; Kuala Lumpur, Malaysia. USA: IEEE; 2010. p. 90–94.  https://doi.org/10.1109/ICSIPA.2009.5478726.
  16. Niemeijer M, Staal J, Ginneken BV, Loog M, Abramoff MD. Comparative study of retinal vessel segmentation methods on a new publicly available database. In: Proc. Society of Photo-Optical Instrumentation Engineers (SPIE 5370); 2004 May 12. Medical Imaging: Image Processing; 2004.  https://doi.org/10.1117/12.535349.
  17. Nugroho HA, Lestari T, Aras RA, Ardiyanto I. Segmentation of retinal blood vessels using Gabor wavelet and morphological reconstruction. In: 2017 3rd International conference on science in information technology (ICSITech). Oct 25–26; Bandung, Indonesia. USA: IEEE; 2018; 2017. p. 513–6.  https://doi.org/10.1109/ICSITech.2017.8257166.CrossRefGoogle Scholar
  18. Razban A, Mahjoory K, Nooshyar M. Segmentation of retinal blood vessels by means of 2D Gabor wavelet and fuzzy mathematical morphology. In: 2016 2nd International conference of signal processing and intelligent systems (ICSPIS). Dec 14–15; Tehran, Iran. USA: IEEE; 2017; 2016. p. 1–5.  https://doi.org/10.1109/ICSPIS.2016.7869877.CrossRefGoogle Scholar
  19. Ruamviboonsuk P, Teerasuwanajak K, Tiensuwan M, Yuttitham K. Interobserver agreement in the interpretation of single-field digital fundus images for diabetic retinopathy screening. Ophthalmology. 2006;113(5):826–32.  https://doi.org/10.1016/j.ophtha.2005.11.021.CrossRefGoogle Scholar
  20. Shahnazi M, Pahlevanzadeh M, Vafadoost M. Wavelet based retinal recognition. In: 2007 9th international symposium on signal processing and its applications. Sharjah, United Arab Emirates. USA: IEEE; 2008; 2007. p. 1–4.  https://doi.org/10.1109/ISSPA.2007.4555369.CrossRefGoogle Scholar
  21. Soares JVB, Leandro JJG, Cesar RM, Jeline HF, Cree MJ. Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification. IEEE Trans Med Imaging. 2006;25(9):1214–22.  https://doi.org/10.1109/TMI.2006.879967.CrossRefGoogle Scholar
  22. Staal JJ, Abramoff MD, Niemeijer M, Viergever MA, van Ginneken Bv. DRIVE: digital retinal images for vessel extraction [Internet]. The Netherlands. 2004 [cited 2018 Jul 20]. Available from: www.isi.uu.nl/Research/Databases/DRIVE/.
  23. Veras RMS, Medeiros FNS, Araújo FHD, Santana AM, Silva RRV. Exudate detection in retina images by mathematical morphology techniques and fuzzy clustering. Rev Bras Eng Biomed. 2013;29(1):45–56.  https://doi.org/10.4322/rbeb.2013.003.CrossRefGoogle Scholar
  24. Waheed Z, Akram UM, Waheed A, Shaukat A. Robust extraction of blood vessels for retinal recognition. In: 2015 Second international conference on information security and cyber forensics (InfoSec); 2015 Nov 15–17; Cape Town, South Africa. USA: IEEE; 2016. p. 1–4.  https://doi.org/10.1109/InfoSec.2015.7435497.
  25. World Health Organization. Blindness and vision impairment prevention - priority eye diseases: Diabetic retinopathy. 2019. https://www.who.int/blindness/causes/priority/en/index5.html. Accessed 26 July 2019.

Copyright information

© Sociedade Brasileira de Engenharia Biomedica 2019

Authors and Affiliations

  1. 1.Graduation Program in Electrical EngineeringPontifical Catholic University of Minas Gerais - PUC-MG, Belo HorizonteBelo HorizonteBrazil
  2. 2.Dr. Ricardo Guimarães Eye HospitalBelo HorizonteBrazil

Personalised recommendations