Title: Machine learning classifiers based on structural ONH measurements for glaucoma diagnosis

Authors: Medha V. Wyawahare; Pradeep M. Patil

Addresses: Vishwakarma Institute of Technology, 666, Upper Indira Nagar, Bibwewadi, Pune, Maharashtra 411037, India ' KJ's Educational Institutes, Kondhwa Saswad Road, Pune, Maharashtra 411048, India

Abstract: Glaucoma which is a leading cause of blindness in the world is not a single disease but a group of disorders with diverse clinical manifestations. If not controlled at an early stage, it causes irreversible damage to vision. Proper medication slows down or even halts its growth. Identifying glaucoma at a very early stage is vital and at the same time difficult. Careful evaluation of optic nerve head structure is extremely important for the disease diagnosis. This paper focuses on use of machine learning classifiers for automatic discrimination between healthy and glaucomatous eyes. Structural optic nerve head measurements in fundus images form the basis of the classification. We investigate and compare performance of eight machine learning classifiers in this work.

Keywords: structural ONH measurements; glaucoma diagnosis; machine learning classifiers; retinal fundus images; optic disc; optic cup; optic nerve head; linear discriminant analysis; LDA; naive Bayes classifier; cup to disc ratio; rim width; blindness.

DOI: 10.1504/IJBET.2016.078338

International Journal of Biomedical Engineering and Technology, 2016 Vol.21 No.4, pp.343 - 360

Received: 24 Jun 2015
Accepted: 02 Dec 2015
Published online: 15 Aug 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article