A Hybrid Image Processing Approach to Examine Abnormality in Retinal Optic Disc

12Citations
Citations of this article
14Readers
Mendeley users who have this article in their library.

Abstract

In ophthalmology, retinal optic nerveand discinspection is a generally adopted procedure to diagnose a variety of diseases, such as aging, diabetes, optic neuritis, glaucoma, papilloedemaand ischaemic optic neuropathy. The paper presents a new method to examine the retina for diseases. Initially, examination of optic disc is considered. During this process, the optic disc recorded using a dedicated digital camera is scrutinized by means of a preferred image processing method in order to mine and appraise the infected section. After receiving the necessary information concerning the disease, the essential treatment procedures are deliberated to limit/cure the disease. This work implements a hybrid approach based on the heuristic algorithm supported multi-level thresholding and image segmentation. This approach is experimentally inspected using the well known RIM-ONE benchmark retinal database. This work implements the pre-processing based on Jaya algorithm guided Shannon's thresholding and post-processing based on the distance regularized level set segmentation. The performance of the proposed segmentation process is then confirmed with the approaches, such as the watershed and Chan-Vese procedures. The extracted optic disc is then compared with the ground truth images provided by the ophthalmologist. The investigational result illustrates that, proposed approach tenders superior average values for image similarity and statistical index compared with the alternatives. This confirms that, proposed hybrid approach can be used in future to evaluate the clinical relevance retinal images.

Cite

CITATION STYLE

APA

Shree, T. D. V., Revanth, K., Raja, N. S. M., & Rajinikanth, V. (2018). A Hybrid Image Processing Approach to Examine Abnormality in Retinal Optic Disc. In Procedia Computer Science (Vol. 125, pp. 157–164). Elsevier B.V. https://doi.org/10.1016/j.procs.2017.12.022

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free