Analysis of hybrid statistical textural and intensity features to discriminate retinal abnormalities through classifiers

Abstract

Retinal image analysis relies on the effectiveness of computational techniques to discriminate various abnormalities in the eye like diabetic retinopathy, macular degeneration and glaucoma. The onset of the disease is often unnoticed in case of glaucoma, the effect of which is felt only at a later stage. Diagnosis of such degenerative diseases warrants early diagnosis and treatment. In this work, performance of statistical and textural features in retinal vessel segmentation is evaluated through classifiers like extreme learning machine, support vector machine and Random Forest. The fundus images are initially preprocessed for any noise reduction, image enhancement and contrast adjustment. The two-dimensional Gabor Wavelets and Partition Clustering is employed on the preprocessed image to extract the blood vessels. Finally, the combined hybrid features comprising statistical textural, intensity and vessel morphological features, extracted from the image, are used to detect glaucomatous abnormality through the classifiers. A crisp decision can be taken depending on the classifying rates of the classifiers. Public databases RIM-ONE and high-resolution fundus and local datasets are used for evaluation with threefold cross validation. The evaluation is based on performance metrics through accuracy, sensitivity and specificity. The evaluation of hybrid features obtained an overall accuracy of 97% when tested using classifiers. The support vector machine classifier is able to achieve an accuracy of 93.33% on high-resolution fundus, 93.8% on RIM-ONE dataset and 95.3% on local dataset. For extreme learning machine classifier, the accuracy is 95.1% on high-resolution fundus, 97.8% on RIM-ONE and 96.8% on local dataset. An accuracy of 94.5% on high-resolution fundus 92.5% on RIM-ONE and 94.2% on local dataset is obtained for the random forest classifier. Validation of the experiment results indicate that the hybrid features can be deployed in supervised classifiers to discriminate retinal abnormalities effectively.

Keywords

Glaucoma fundus image blood vessel thresholding clustering feature selection classifier

Get full access to this article

View all access options for this article.

References

World Health Organization. Prevention of blindness visual impairment, www.who.int/blindness/causes/priority/en

Tham

Wong

et al . Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis. Ophthalmology 2014; 121: 2081–2090.

Greaney

Hoffman

Garway-Heath

et al . Comparison of optic nerve imaging methods to distinguish normal eyes from those with glaucoma. Investig Ophthalmol Vis Sci 2002; 43: 140–145.

Shen

Wong

Foster

et al . Prevalence and types of glaucoma in Malay people: the Singapore Malay eye study. Investig Ophthalmol Vis Sci 2008; 49: 2846–3851.

Bressler

Fine

. Age-related macular degeneration. Surv Ophthalmol 1988; 32(6): 375–413.

Ramani

Lakshmi

. Automatic retinal vessel segmentation through Gabor filtering, principal component analysis and ensemble of classifiers. Adv Nat Appl Sci 2015; 9: 600–607.

Abramoff

Garvin

Sonka

. Retinal imaging and image analysis. IEEE Rev Biomed Eng 2010; 3: 169–208.

Fraz

Remagnino

Hoppe

et al . Blood vessel segmentation methodologies in retinal images—a survey. Comput Meth Prog Biomed 2012; 1: 407–433.

Swindale

Stjepanovic

Chin

et al . Automated analysis of normal and glaucomatous optic nerve head topography images. Investig Ophthalmol Vis Sci 2000; 41(7): 1730–1742.

10.

Krig

. Image pre-processing. In: Krig

(ed.) Computer vision metrics. Berlin: Springer, 2014, pp.39–83.

11.

Cinsdikici

Aydin

. Retinal vessel segmentation in ophthalmoscope images using matched filters. Comput Meth Prog Biomed 2009; 96: 85–95.

12.

Vermeer

Vos

Lemij

et al . A model based method for retinal blood vessel detection. Comput Biol Med 2004; 34: 209–219.

13.

Marinez-Perez

Hughes

Thom

et al . Segmentation of blood vessels from red-free and fluorescein retinal images. Med Image Anal 2007; 11: 47–61.

14.

Jumb

Sohani

Shrinivas

. Color image segmentation using K-means clustering and Otsu’s thresholding. Int J Innovat Technol 2014; 3: 72–76.

15.

Acharya

EYK

Eugene

LWJ

et al . Decision support system for the glaucoma using Gabor transformation. Biomed Sig Process Control 2015; 15: 18–26.

16.

Dua

Rajendra Acharya

Chowriappa

et al . Wavelet-based energy features for glaucoma image classification. IEEE T Inf Technol Biomed 2012; 16(1): 80–87.

17.

Yadav

Partha Sarathi

Dutta

. Classification of glaucoma based on texture features using neural networks. In: Proceedings of 7th international conference on contemporary computing, Noida, India, 7–9 August 2014. New York: IEEE.

18.

Hoover

Kouznetsova

Goldbaum

. Locating blood vessels in retinal images by piecewise thresholding probing of a matched filter response. IEEE T Med Imaging 2000; 19: 203–211.

19.

Khalil

Khalid

Syed

. Review of machine learning techniques for glaucoma detection and prediction. In: Science and information conference, London, 27–29 August 2014, pp.27–29. New York: IEEE.

20.

Salazar-Gonzalez

Kaba

et al . Segmentation of the blood vessels and optic disc in retinal images. IEEE J Biomed Health Inf 2014; 18: 2168–2194.

21.

Tolias

Panas

. A fuzzy vessel tracking algorithm for retinal images based on fuzzy clustering. IEEE T Med Imaging 1998; 17(2): 263–273.

22.

Neimaijer

Staal

van Ginneken

et al . Comparative study of retinal vessel segmentation methods on a new publicly available database. In: Proceedings of SPIE5370: medical imaging, image processing, San Diego, CA, 12 May 2004, pp.648–656. Bellingham, WA: SPIE.

23.

Staal

Abramoff

Niemeijer

et al . Ridge-based vessel segmentation in color images of the retina. IEEE T Image Process 2004; 23: 501–509.

24.

Sinthanayothin

Boyce

Cook

et al . Automated localization of optic disk, fovea, and retinal blood vessels from digital color fundus images. Br J Ophthalmol 1999; 83(9): 902–910.

25.

Gardner

Keating

Williamson

et al . Automatic detection of diabetic retinopathy using an artificial neural network. Br J Ophthalmol 1996; 80(11): 940–944.

26.

Walter

Massin

Erginay

et al . Automatic detection of microneurysms in color fundus images. Med Image Anal 2007; 11(6): 555–566.

27.

Chaudhuri

Chatterjee

Katz

et al . Detection of blood vessels in retinal images using 2 D matched filters. IEEE T Med Imaging 1989; 8: 263–269.

28.

Soares

JVB

Leandro

JJG

Cesar

Jr et al . Retinal blood segmentation using 2D Gabor wavelet and supervised classification. IEEE T Med Imaging 1006; 25: 1214–1222.

29.

Manjiri

Ramesh

Yogesh

et al . Automated localization of optic disc, detection of microneurysms and extraction of blood vessels to bypass angiography. In: Satapathy

Biswal

Udgata

et al . (eds) Proceedings of the 3rd international conference on frontiers of intelligent computing: theory and applications. Cham: Springer, 2015, pp.579–587.

30.

Haralick

Shanmugam

Dinstein

. Textural features for image classification. IEEE T Syst Man Cy 1973; 3(6): 610–621.

31.

Yoo

Ackerman

Lorensen

et al . Engineering and algorithm design for an image processing Api: a technical report on ITK—the insight toolkit. In: Westwood

(ed.) Medicine meets virtual reality. Amsterdam: IOS Press, 2002, pp.586–592.

32.

Nitish

Vrushsen

. GLCM textural features for brain tumor classification. IJCSI Int J Comput Sci 2012; 9(3): 354–359.

33.

Ojala

Pietikäinen

Harwood

et al . A comparative study of texture measures with classification based on featured distributions. Pattern Recogn 1996; 29(1): 51–59.

34.

Guo

Zhang

et al . A completed modeling of local binary pattern operator for texture classification. IEEE T Image Proces 2010; 19(6): 1657–1663.

35.

Noah

Bob

. Color Moments 2005; 0341091.

36.

Mookiah

MRK

Acharya

Martis

et al . Evolutionary algorithm based classifier parameter tuning for automatic diabetic retinopathy grading: a hybrid feature extraction approach. Knowl-Based Syst 2012; 39: 9–22.

37.

Vapnik

. Statistical learning theory. New York: Wiley, 1998.

38.

Srivastava

Bhambhu

. Data classification using support vector machine. J Theor Appl Inf Technol 2009; 12.

39.

Kotsiantis

. Supervised machine learning: a review of classification techniques. Informatica 2007; 31: 249–268.

40.

Huang

Zhu

Siew

et al . Extreme learning machine: theory and applications. Neurocomputing 2006; 70(1): 489–501.

41.

Huang

Zhou

Ding

et al . Extreme learning machine for regression and multiclass classification. IEEE T Syst Man Cy B 2012; 42: 513–529.

42.

Huang

Chen

. Convex incremental extreme learning machine. Neurocomputing 2007; 70: 3056–3062.

43.

Suresh

Babu

Kim

. No-reference quality assessment using modified extreme learning machine classifier. Appl Soft Comput 2009; 9: 541–552.

44.

Breimen

. Random forests. Mach Learn 2001; 45: 5–32.

45.

https://www5.cs.fau.de/research/data/fundus-images

46.

Akram

Tariq

. Glaucoma detection using novel optic disc localization, hybrid feature set and classification technique. Australas Phys Eng Sci Med 2015; 38: 643–655.

47.

Fumero

Alayon

Sanchez

et al . RIM-ONE: an open retinal image database for optic nerve evaluation. In: International symposium on computer based medical system, Bristol, 27–30 June 2011, Vol. 1, pp.1–6. New York: IEEE.

48.

Nayak

Rajendra Acharya

Subbanna Bhat

et al . Automated diagnosis of glaucoma using digital fundus images. Berlin: Springer, 2008, pp.337–346.

49.

Kim

Iftekharuddin

Davey

et al . Novel fractal features-based multiclass glaucoma detection and progression prediction. IEEE T Biomed Health Inf 2013; 17: 269–276.

50.

Septiarini

Hamdani Khairinia

. The contour extraction of cup in fundus images for glaucoma detection. Int J Electr Comput Eng 2016; 6: 2797–2804.

51.

Karthikeyan

Rengarajan

. Performance analysis of gray level co-occurrence matrix features for glaucoma diagnosis. Am J Appl Sci 2014; 11: 248–257.

52.

Rajendra Acharya

Dua

et al . Automated diagnosis of Glaucoma using texture and higher order spectra features. IEEE T Inf Tech. Biomed 2011; 15(3): 449–455.