Automated Detection of Optic Disc and Exudates in Retinal Images

Digital colour retinal images are used by the ophthalmologists for the detection of many eye related diseases such as Diabetic retinopathy. These images are generated in large number during the mass screening of the disease and may result in biased observation due to fatigue. Automated retinal image processing system could save workload of the ophthalmologists and also assist them to extract the normal and abnormal structures in retinal images and help in grading the severity of the disease. In this paper we present a method for automatic detection of optic disc followed by classification of hard exudates pixels in retinal image. Optic disc localization is achieved by iterative threshold method to identify initial set of candidate regions followed by connected component analysis to locate the actual optic disc. Exudates are detected using k means clustering algorithm. The algorithm is evaluated against a carefully selected database of 100 color retinal images at different stages of diabetic retinopathy. The methods achieve a sensitivity of 92% for the optic disc and 86% for the detection of exudates.

[1]  C. Sinthanayothin,et al.  Automated localisation of the optic disc, fovea, and retinal blood vessels from digital colour fundus images , 1999, The British journal of ophthalmology.

[2]  Enrico Grisan,et al.  Detection of optic disc in retinal images by means of a geometrical model of vessel structure , 2004, IEEE Transactions on Medical Imaging.

[3]  Shankar M. Krishnan,et al.  Automatic image analysis of fundus photograph , 1997, Proceedings of the 19th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 'Magnificent Milestones and Emerging Opportunities in Medical Engineering' (Cat. No.97CH36136).

[4]  T. Williamson,et al.  Automatic detection of diabetic retinopathy using an artificial neural network: a screening tool. , 1996, The British journal of ophthalmology.

[5]  Huiqi Li,et al.  Automated feature extraction in color retinal images by a model based approach , 2004, IEEE Transactions on Biomedical Engineering.