Study on Computational Intelligence Based Image Processing Techniques for Detection of Retinal Diseases in Fundus Images

Abstract

The colour fundus images of human retina have been enhanced and utilized by the medical practitioner for the diagnosis of various eye-related diseases like proliferative and non-proliferative diabetic retinopathy, diabetic maculopathy and glaucoma. Due to the increase in population and scarceness of the ophthalmologist, screening more number of patients becomes a significant task. Moreover, the manual process is tedious, time consuming and impractical. For large data to be handled, there is a need for an automatic processing system that helps ophthalmologists. This great challenge opens up the possibility of developing algorithms with the recent technological advancements towards automation of screening fundus images. The image processing algorithms employed on fundus images play an important role in detection of pathologies, which improves the detection system in diverse angles. The computation-based detection system reduces human error and thus gives accurate result and is helpful for the ophthalmologist in the direction of the right treatment for patients. The customary screening process of fundus images is essential to provide appropriate and timely treatment for patients to reduce the incidence of impaired vision and blindness. Many researchers are performing study and exploring to make the diagnosis system more effective by increasing the accuracy of the system newline