Peer Reviewed

1

DOI

10.17077/omia.1004

Conference Location

Boston, MA, USA

Publication Date

September 2014

Abstract

We develop a general framework for multiple ocular diseases diagnosis, based on Graph Regularized Multi-label Learning (GRML). Glaucoma, Pathological Myopia (PM), and Age-related Macular Degeneration (AMD) are three leading ocular diseases in the world. By exploiting the correlations among these three diseases, a novel GRML scheme is investigated for a simultaneous detection of these three leading ocular diseases for a given fundus image. We validate our GRML framework by conducting extensive experiments on SiMES dataset. The results show area under curve (AUC) of the receiver operating characteristic curve in multiple ocular diseases detection are much better than traditional popular algorithms. The method could be used for glaucoma, PM, and AMD diagnosis.

Rights

Copyright © 2014, Xiangyu Chen, Yanwu Xu, Lixin Duan, Zhuo Zhang, Damon Wing Kee Wong, and Jiang Liu.

Included in

Ophthalmology Commons

Share

COinS
 
Sep 14th, 12:00 AM Sep 14th, 12:00 AM

Multiple ocular diseases detection by graph regularized multi-label learning

Boston, MA, USA

We develop a general framework for multiple ocular diseases diagnosis, based on Graph Regularized Multi-label Learning (GRML). Glaucoma, Pathological Myopia (PM), and Age-related Macular Degeneration (AMD) are three leading ocular diseases in the world. By exploiting the correlations among these three diseases, a novel GRML scheme is investigated for a simultaneous detection of these three leading ocular diseases for a given fundus image. We validate our GRML framework by conducting extensive experiments on SiMES dataset. The results show area under curve (AUC) of the receiver operating characteristic curve in multiple ocular diseases detection are much better than traditional popular algorithms. The method could be used for glaucoma, PM, and AMD diagnosis.