Page 221 - Computational Retinal Image Analysis

References 217

4 Conclusions

Optic disc and optic cup segmentation are important steps for glaucoma detection. In
the past, many methods have been proposed to segment the optic disc and optic cup for
glaucoma detection. Many deep learning-based methods have been proposed as well
recently. However, most of these methods neglect image quality, an important factor
that affects the performance of retinal image analysis. In this chapter, we introduced
a novel SGRIF technique to remove artifacts caused by cataractous lens, a frequent
cause of limited image quality. Experimental results show that the method enhances
the contrast of retinal images, as measured by HFM, HS, and VLL. The technique is
further validated with retinal image analysis tasks including optic cup segmentation
and CDR measurement. Both experiments show that the method improves the
accuracy of the tasks results. In this chapter, we mainly apply the algorithm in the
optic disc region and not for optic disc segmentation as the optic disc boundary is
usually much stronger and will not be smoothed away in GIF. Therefore, the benefit
from our method is less obvious. Since our method improves the image quality in
the optic disc area, it could be used to improve automatic analysis for disc-related
disease detection. Our results also suggest, qualitatively, that the method enhances
the blood vessel. In future work, we will also explore how it affects other analysis
tasks such as vessel detection and lesion detection quantitatively.

References
[1] H.A. Quigley, A.T. Broman, The number of people with glaucoma worldwide in 2010
and 2020, Br. J. Ophthalmol. 90 (3) (2006) 262–267.
[2] S.Y. Shen, T.Y. Wong, P.J. Foster, J.L. Loo, M. Rosman, S.C. Loon, W.L. Wong, S.M. Saw,
T. Aung, The prevalence and types of glaucoma in Malay people: the Singapore Malay
eye study, Invest. Ophthalmol. Vis. Sci. 49 (9) (2008) 3846–3851.
[3] P.J. Foster, F.T. Oen, D. Machin, T.P. Ng, J.G. Devereux, G.J. Johnson, P.T. Khaw,
S.K. Seah, The prevalence of glaucoma in Chinese residents of Singapore: a cross-
sectional population survey of the Tanjong Pagar district, Arch. Ophthalmol. 118 (8)
(2000) 1105–1111.
[4] M.D. Abràmoff, M.K. Garvin, M. Sonka, Retinal imaging and image analysis, IEEE
Trans. Med. Imaging 3 (2010) 169–208.
[5] H. Li, O. Chutatape, Boundary detection of optic disc by a modified ASM method,
Pattern Recogn. 36 (2003) 2093–2104.
[6] A. Aquino, M. Gegundez-Arias, D. Marin, Detecting the optic disc boundary in digital
fundus images using morphological, edge detection, and feature extraction techniques,
IEEE Trans. Med. Imaging 29 (2010) 1860–1869.
[7] C. Muramatsu, T. Nakagawa, A. Sawada, Y. Hatanaka, T. Hara, T. Yamamoto, H. Fujita,
Automated segmentation of optic disc region on retinal fundus photographs: comparison
of contour modeling and pixel classification methods, Comput. Methods Prog. Biomed.
101 (2011) 23–32.
[8] J. Cheng, J. Liu, D.W.K. Wong, F. Yin, C. Cheung, M. Baskaran, T. Aung, T.Y. Wong,