Retinal image analysis for diagnosis of macular edema using digital fundus images

Digital fundus images are one of the modern and advanced approaches of creating image of inner surface of human eye emphasizing retina. These fundus images are really helpful in diagnosis of possible abnormalities and severe diseases like diabetic macular edema and its various types. Research has shown that early detection and treatment can prevent total vision loss and severe impacts on human visual system. Hence an automated system for diagnosing macular edema will help the ophthalmologists and patients. In this paper, we have proposed a novel method for diagnosing macular edema using fundus images. The technique has four steps which constitutes of preprocessing, macula detection, feature extraction of possible exudates region followed by classification using Naïve Bayes classifier. The proposed system is tested using MESSIDOR database and results show that our method outperformed others in terms of accuracy.