AUTOMATIC DIAGNOSIS OF FAMILIAL EXUDATIVE VITREORETINOPATHY

Abstract: A snapshot of retinal image is used to analyse the disease called familial exudative vitreoretinopathy (FEVR). FEVR disease mostly affects the retinal nerve parts and it leads to vision loss, retinal detachment, strabismus, and a visible whiteness (leukocoria) in the normally black pupil. The symptoms may vary even within the same family. This disease is incurable when it reaches its severe stage. So it is very important to diagnose it in previous stage of infection. Along with FEVR we also diagnose the disease like glaucoma, refractive power and cataract. Mostly diabetes patients are affected with such type of retinal disease. Automatic retinal segmentation is complicated by the fact that retinal images are often noisy, poorly contrasted, and the vessel widths can vary from very large to very small. So in this project, we implement automate segmentation approach based on graph theoretical method to provide regional information using measure. We represent the segmented vascular structure of retina as a vessel segment graph and make problem of identify the vessels as one of finding the blood vessels to have good correlation. We plan a method of image processing with some insisted algorithms to diagnose and evaluate the retinal disease.

Keywords: Image Processing, SVM algorithm, IPACHI Model, MATLAB.