Improved automatic localization of optic disc in Retinal Fundus using image enhancement techniques and SVM

Fundus Image is the creation of a photograph of human eye interior surface, including retina, optic disc, macula, and posterior pole. Fundus photography is used by ophthalmologist's and trained medical professionals for monitoring progression of a disease, diagnosis of a disease or in screening programs and epidemiology. The optic disc is also the entry point for the major blood vessels that supply the retina. The detection and localization of optic disc from ophthalmic fundus images is very vital for the diagnosis of various diseases in ophthalmology. These diseases are varying from Glaucoma to Diabetic Retinopathy and many more. Image processing is widely used for the purpose of diagnosing and detection of different retinal diseases however, the results generated by these algorithms can only be tested by domain experts. This research aims at an image processing method for detecting optic disc. An algorithm is developed and generated results are verified by image processing algorithm for detection of optic disc using machine learning through Support Vector Machine (SVM). The algorithm is first trained on a set of images as marked by the domain expert (Ophthalmologist) and SVM training results are stored in a database. Image processing technique is used for the testing of unknown fundus image and its outcome is verified through SVM. Depending on SVM outcome, desired optic disc is marked on the image. The proposed framework uses image processing algorithm and then verify those results using SVM. The proposed method is tested on a known public repository MESSIDOR and STARE Fundus image database and applied accomplished results to 100 images.