Paper Info
Title
An Approach For Glaucoma Detection Based On The Features Representation In Radon Domain
Abstract
Glaucoma is a chronic and irreversible eye disease that leads to the structural changes of the Optic Nerve Head (ONH). In clinical practice, ONH assessment is one of the most significant measurements for glaucoma detection. However, the structural changes of ONH reveals complex mixture of visual patterns that are challenging to be represented. In this paper, a novel features representation approach in Radon domain is proposed to capture these complex patterns. In our method, fundus images are projected into Radon domain with Radon Transform (RT) in which the spatial radial variations of ONH are converted to a discrete signal for constraint optimization, feature enhancement and dimensionality reduction. Subsequently, the Discrete Wavelet Transform (DWT) is adopted to obtain subtle differences and quantize them. The experiments show that our approach achieves excellent detection results on RIMONE-r2 dataset with the accuracy and Area Under the Curve (AUC) of receiver operating characteristic curve at 86.154% and 0.906 respectively, much better than other algorithms. The results demonstrate that the proposed method can be used as an effective tool for glaucoma detection in the mass screening of fundus images.
Year
DOI
Venue
2018
10.1007/978-3-319-95933-7_32
INTELLIGENT COMPUTING THEORIES AND APPLICATION, PT II
Keywords
Field
DocType
Computer-aided diagnosis, Glaucoma detection, Radon transform
Glaucoma,Receiver operating characteristic,Dimensionality reduction,Discrete-time signal,Pattern recognition,Computer science,Computer-aided diagnosis,Fundus (eye),Artificial intelligence,Discrete wavelet transform,Radon transform
Conference
Volume
ISSN
Citations 
10955
0302-9743
0
PageRank 
References 
Authors
0.34
4
6
Name
Order
Citations
PageRank
Beiji Zou123141.61
Qilin Chen200.68
Rongchang Zhao393.81
Ping-Bo Ouyang432.77
Chengzhang Zhu511.38
Xuanchu Duan621.03