Automatic Microaneurysm Quantification for Diabetic Retinopathy Screening

Authors: A. Sopharak, B. Uyyanonvara, S. Barman

Abstract:

Microaneurysm is a key indicator of diabetic retinopathy that can potentially cause damage to retina. Early detection and automatic quantification are the keys to prevent further damage. In this paper, which focuses on automatic microaneurysm detection in images acquired through non-dilated pupils, we present a series of experiments on feature selection and automatic microaneurysm pixel classification. We found that the best feature set is a combination of 10 features: the pixel-s intensity of shade corrected image, the pixel hue, the standard deviation of shade corrected image, DoG4, the area of the candidate MA, the perimeter of the candidate MA, the eccentricity of the candidate MA, the circularity of the candidate MA, the mean intensity of the candidate MA on shade corrected image and the ratio of the major axis length and minor length of the candidate MA. The overall sensitivity, specificity, precision, and accuracy are 84.82%, 99.99%, 89.01%, and 99.99%, respectively.

Keywords: Diabetic retinopathy, microaneurysm, naive Bayes classifier

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1084604

References:

[1] S. Wild, G. Roglic, A. Green et al., "Global prevalence of diabetes: estimates for the year 2000 and projections for 2030," Diabetes Care 27, 2004, pp.1047-1053.
[2] P. Massin, A. Erginay, and A. Gaudric, "Retinopathie Diabetique", Elsevier, Editions scientifiques of medicales, Elsevier, SAS, Paris 2000.
[3] T. Spencer, J.A. Olson, K.C. McHardy et al., "An imageprocessingstrategy for the segmentation and quantification of microaneurysms in fluorescein angiograms of the ocular fundus," Comp Biomed Res 29, 1996, pp. 284-302.
[4] M.J. Cree, J.A. Olson, K.C. McHardy et al., "A fully automated comparative microaneurysm digital detection system," Eye 11, 1997, pp. 622-628.
[5] A. Frame, P. Undrill, M. Cree et al., "A comparison of computer based classification methods applied to the detection of microaneurysms in ophthalmic fluorescein angiograms," Comput. Biol. Med. 28, 1998, pp. 225-238.
[6] T. Walter, P. Massin, A. Erginay et al., "Automatic detection of microaneurysms in color fundus images," Medical Image Analysis 11(6), 2007, pp.555-566.
[7] C. Sinthanayothin, J.F. Boyce, T.H Williamson, T.H. et al., "Automated Detection of Diabetic Retinopathy on Digital Fundus Image," Diabetic Medicine 19(2), 2002, pp. 105-112, 2002.
[8] D. Usher, M. Dumskyj, M. Himaga et al., "Automated Detection of Diabetic Retinopathy in Digital Retinal Images: A Tool for Diabetic Retinopathy Screening," Diabetic Medicine 21(1), 2004, pp. 84-90.
[9] B. Dupas, T. Walter, A. Erginay et al., "Evaluation of automated fundus photograph analysis algorithms for detecting microaneurysms, haemorrhages and exudates, and of a computer-assisted diagnostic system for grading diabetic retinopathy," Diabetes & Metabolism 36(3), 2010, pp. 213-220.
[10] M. Niemeijer, B. van Ginneken, J. Staal et al., "Automatic detection of red lesions in digital color fundus photographs," IEEE Trans Med Imaging 24(5), 2005, pp.584-592.
[11] G. Gardner, D. Keating, T.H. Williamson et al., " Automatic detection of diabetic retinopathy using an artificial neural network: a screening tool," Br J Ophthalmol 80, 1996, pp.940-944.
[12] B. Zhang, X. Wu, J. You et al., "Detection of microaneurysms using multi-scale correlation coefficients," Pattern Recognition 43(6), 2010, pp. 2237-2248.
[13] A.Sopharak, B. Uyyanonvara, and S. Barman, Automatic Microaneurysm Detection from Non-dilated Diabetic Retinopathy Retinal Images Using Mathematical Morphological Methods. IAENG International Journal of Computer Science 38(3) (2011), 295-301.
[14] A. Sopharak, B. Uyyanonvara, S. Barman et al., "Automatic detection of diabetic retinopathy exudates from non-dilated retinal images using mathematical morphology methods," Computer Medical Imaging and Graphics 32(8), 2008, pp. 720-727.
[15] R.S. Kenneth, C.R. John, J. Matthew et al. (2006, June 15). Difference of Gaussians Edge Enhancement (Online). Available: http://micro.magnet.fsu.edu/primer/java/digitalimaging/processing/diffga ussians/index.html
[16] N. Friedman, D. Geiger, and M. Goldszmidt, "Bayesian network classifiers," Machine Learning. Vol. 29, pp.131-163, 1997.
[17] O.D. Richard, E.H. Peter, and G.S. David, Pattern Classification 2nd edition, A Wiley-Interscience Publication, 2000, pp. 20-83.
[18] X.Y. Wang, J. Garibaldi, and T. Ozen, "Application of The Fuzzy CMeans clustering Method on the Analysis of non Pre-processed FTIR Data for Cancer Diagnosis," Internat. Conf. on Australian and New Zealand Intelligent Information Systems (ANZIIS), pp. 233-238, 2003.