Denetimli Sınıflandırma Yöntemleri ile Retinal Kan Damarı Bölütleme

Öz

Retinal fundus görüntülerde kan damarı bölütleme işlemi, diyabetik retinopati, glukoma gibi bazı hastalıkların teşhisi ve ön tanısı için önemli bir aşamadır. Bu çalışmada renkli retinal fundus görüntülerde damar bölütleme amacıyla kullanılan denetimli sınıflandırma yöntemleri uygulamalı olarak karşılaştırılmaktadır. Sınıflandırma işleminden önce kan damarı ve retina arkaplan piksellerini birbirinden ayıracak şekilde damar iyileştirmeye dayalı piksel tabanlı özellik çıkarma işlemi gerçekleştirilir. Daha sonra çıkarılan bu özellikler kullanılarak sınıflandırıcı yardımıyla piksellerin kan damarına ya da arkaplana ait olup olmadığına karar verilir. Denetimli sınıflandırma yöntemi olarak k en yakın komşuluk, Naive Bayes sınıflandırıcı ve destek vektör makinaları kullanılmaktadır. Performans değerlendirmesi için internet üzerinde erişilebilir olan STARE ve DRIVE veritabanları kullanılmaktadır. Sonuç olarak elde edilen başarım değerleri ve işlem süreleri karşılaştırılmıştır. Naive Bayes sınıflandırma yönteminin en hızlı ve destek vektör makinarı yönteminin ise diğerlerine göre daha yüksek başarı sağladığı gözlenmiştir.

Anahtar Kelimeler

• Denetimli sınıflandırma, k en yakın komşular, Naive Bayes, destek vektör makineleri, öznitelik çıkarma, damar bölütleme, kan damarı iyileştirme

Retinal Blood Vessel Extraction with Supervised Classification Methods

Öz

Blood vessel segmentation in retinal fundus images is the first step for the diagnosis and treatment of diseases such as diabetic retinopathy, glaucoma and age related macular degeneration. In this paper, several supervised classification methods with adapted features are used in order to extract blood vessels in color retinal fundus images. Furthermore, the obtained results are compared against each other in terms of computational durations and classification accuracy. Firstly, a pixel based feature extraction method is performed in which features are extracted from the enhanced images of the vessel. Afterwards, a classification stage is performed to decide whether a pixel belongs to a vessel or the retinal background using these features. K-nearest neighbors, Naïve Bayes and support vector machines are used as supervised classification mechanisms. Retinal fundus images from two publicly available database STARE and DRIVE are used for performance evaluation. Obtained performance values and computation time results are compared. As a result, it is observed that Naïve Bayes classifier is the fastest method and support vector machines method has the highest accuracy.

Anahtar Kelimeler

• Supervised classification, k nearest neighbors, Naive Bayes, support vector machines, feature extraction, blood vessel enhancement, vessel segmentation

Kaynakça

1. S. Chaudhuri, S. Chatterjee, N. Katz, M. Nelson, and M. Goldbaum, “Detection of blood vessels in retinal images using two-dimensional matched filters”, IEEE Trans. Med. Imaging, vol. 8, pp. 263–269, 1989.
2. M. M. Fraz, P. Remagnino, A. Hoppe, B. Uyyanonvara, a R. Rudnicka, C. G. Owen, and S. a Barman, “Blood vessel segmentation methodologies in retinal images-a survey”, Comput. Methods Programs Biomed., vol. 108, no. 1, pp. 407–433, Oct. 2012.
3. J. Staal, M. D. Abràmoff, M. Niemeijer, M. A. Viergever, and B. Van Ginneken, “Ridge-based vessel segmentation in color images of the retina”, IEEE Trans. Med. Imaging, vol. 23, no. 4, pp. 501–509, 2004.
4. J. V. B. Soares, J. J. G. Leandro, R. M. Cesar Jşnior, H. F. Jelinek, and M. J. Cree, “Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification”, IEEE Trans. Med. Imaging, vol. 25, no. 9, pp. 1214– 1222, Sep. 2006.
5. E. Ricci and R. Perfetti, “Retinal blood vessel segmentation using line operators and support vector classification”, IEEE Trans. Med. Imaging, vol. 26, no. 10, pp. 1357–1365, Oct. 2007.
6. D. Marín, A. Aquino, M. E. Gegşndez-Arias, and J. M. Bravo, “A new supervised method for blood vessel segmentation in retinal images by using gray-level and moment invariants-based features”, IEEE Trans. Med. Imaging, vol. 30, no. 1, pp. 146–158, 2011.
7. A. Bhuiyan, B. Nath, J. Chua, and R. Kotagiri, “Blood vessel segmentation from color retinal images classification” in Image Processing, 2007. ICIP 2007. IEEE International Conference on, 2007, vol. 5, pp. 521–524. texture
8. A. Hoover, V. Kouznetsova, and M. Goldbaum, “Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response”, IEEE Trans. Med. Imaging, vol. 19, no. 3, pp. 203–210, 2000.

9. Z. Yavuz and C. Kose, “Blood vessel segmentation from retinal images based on enhencement methods” in Signal Processing and Communications Applications Conference (SIU), 2014 22nd, 2014, pp. 907–910.
10. M. M. Fraz, A. Basit, P. Remagnino, A. Hoppe, and S. A. Barman, “Retinal vasculature segmentation by morphological curvature, reconstruction thresholding”, in 2011 7th International Conference on Emerging Technologies, 2011, pp. 1–6. hysteresis
11. A. M. Mendonça and A. Campilho, “Segmentation of retinal blood vessels by combining the detection of centerlines and morphological reconstruction”, IEEE Trans. Med. Imaging, vol. 25, no. 9, pp. 1200–13, Sep. 2006.
12. O. Chutatape, L. Zheng, and S. M. Krishnan, “Retinal blood vessel detection and tracking by matched Gaussian and Kalman filters” in Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE, 1998, vol. 6, pp. 3144–3149.
13. F. K. H. Quek and C. Kirbas, “Vessel extraction in medical images by wave-propagation and traceback”, Med. Imaging, IEEE Trans., vol. 20, no. 2, pp. 117–131, 2001.
14. M. Frangi, AlejandroF. and Niessen, WiroJ. and Vincken, KoenL. and Viergever, “Multiscale vessel enhancement filtering” in Medical Image Computing and Computer-Assisted Interventation — MICCAI’98, 1998, vol. 1496, pp. 130–137.
15. M. E. Martinez-Perez, A. D. Hughes, S. a Thom, A. a Bharath, and K. H. Parker, “Segmentation of blood vessels from red-free and fluorescein retinal images”, Med. Image Anal., vol. 11, no. 1, pp. 47–61, Feb. 2007.
16. L. Espona, M. Carreira, M. Ortega, and M. Penedo, “A snake for retinal vessel segmentation”, Pattern Recognition and Image Analysis, pp. 178–185, 2007.
17. B. Al-Diri, A. Hunter, and D. Steel, “An active contour model for segmenting and measuring retinal vessels”, IEEE Trans. Med. Imaging, vol. 28, no. 9, pp. 1488–97, Sep. 2009.
18. A. Hoover, “STructured Analysis of the REtina (STARE)”, http://www.ces.clemson.edu/~ahoover/stare/.
19. J. Staal, “DRIVE: Digital Retinal Images for Vessel Extraction”, [Online]. Available: http://www.isi.uu.nl/Research/Databases/DRIV E/.
20. Z. Yavuz and C. Kose, “Retinal Blood Vessel Segmentation Using Gabor Filter and Top-Hat Transform” in 19th Conference on Signal Processing and Communications Applications (SIU 2011), 2011, pp. 546–549.
21. S. M. Pizer, E. P. Amburn, J. D. Austin, R. Cromartie, A. Geselowitz, T. Greer, B. ter Haar Romeny, J. B. Zimmerman, and K. Zuiderveld, “Adaptive histogram equalization and its variations”, Comput. vision, Graph. image Process., vol. 39, no. 3, pp. 355–368, 1987.
22. Hand, David J., and Keming Yu. "Idiot's Bayes—not so stupid after all?."International statistical review 69.3 (2001): 385-398.
23. V. Vapnik, I. Guyon, and T. Hastie, “Support vector machines”, 1995. [Online]. Available: http://www.support-vector-machines.org. [Accessed: 10-Jan-2014].