A novel focus for segmentation chained with property extraction and classification of branch retinal artery occlusion an ophthalmic examination

Abstract

Branch Retinal Artery Occlusion (BRAO) has become a serious newlinedisorder causing permanent loss of vision. It is mainly caused due to the newlinerupture of blood vessels which is too small in size and thereby increases the newlinerisk for efficient detection. Therefore an efficient detection scheme is newlinenecessary to evaluate appropriately and pave the way for systemic therapy to newlinepreserve or recover vision in the affected eye. High rated and active research newlinein Ophthalmic field has paved the way for the detection of several retinal newlinedisorders which helps the ophthalmologist, in scheduling and executing newlinetimely treatment. There exist different imaging device for retinal examination newlinein ophthalmology with varying degrees of resolution. The images obtained newlinefrom such imaging devices contains noise which must be eliminated for newlineproducing accurate results. Moreover, BRAO is the most serious and rare case newlinein ophthalmology for which the automatic segmentation of BRAO lesions is newlineattracted by several researchers. BRAO is mainly caused due to occlusion in newlinethe arteries. BRAO are mainly noticed in elder persons and are rare in newlineyounger generation.Chapter 2 reviews the literatures of various systems newlineavailable in the previous work for differentiating the abnormal and normal newlinecases. Collecting the database for BRAO was another challenging task for newlinethis research work since there could be only one or two BRAO patients per newlinehospital. Thus in this research work, 100 retinal images are processed newline(Clinical and public retinal database images ie. retinal image bank for newlineabnormal images and Dr Hossein rabbani datasets for normal images).These newlineimages are generated, arranged and stored during data collection. The newlinecollected images are initially converted into JPEG or JPG format. newline newline