Automatic localization of the anatomic structures in retinal fundus images

Abstract

The work in this thesis mainly focuses on the development of an automatic system for the purpose of detecting anatomical features such as blood vessel optic disc and fovea in colour retinal fundus images The detection of these features is important prerequisite for the diagnosis of diabetic related eye diseases such as diabetic retinopathy gluacoma and diabetic macular edema Early diagnosis and treatment of these diseases are very important to prevent it from getting worse eventually leading to blindness Conventional approaches for early diagnosis of these diseases are inefficient that they not only waste ophthalmologists time but are also expensive for the patients In India there are not enough resources in terms of time and available expert ophthalmologists for carrying on an extensive screening Therefore computer aided diagnosis is important for the early and efficient diagnosis of eye diseases newline