Deep convolution neural network for accurate diagnosis of glaucoma using digital fundus images

U. Raghavendra, Hamido Fujita, Sulatha V. Bhandary, Anjan Gudigar, Jen Hong Tan, U. Rajendra Acharya

Research output: Contribution to journalArticle

59 Citations (Scopus)

Abstract

Glaucoma progressively affects the optic nerve and may cause partial or complete vision loss. Raised intravascular pressure is the only factor which can be modified to prevent blindness from this condition. Accurate early detection and continuous screening may prevent the vision loss. Computer aided diagnosis (CAD) is a non-invasive technique which can detect the glaucoma in its early stage using digital fundus images. Developing such a system require diverse huge database in order to reach optimum performance. This paper proposes a novel CAD tool for the accurate detection of glaucoma using deep learning technique. An eighteen layer convolutional neural networks (CNN) is effectively trained in order to extract robust features from the digital fundus images. Finally these features are classified into normal and glaucoma classes during testing. We have achieved the highest accuracy of 98.13% using 1426 (589: normal and 837: glaucoma) fundus images. Our experimental results demonstrates the robustness of the system, which can be used as a supplementary tool for the clinicians to validate their decisions.

Original languageEnglish
Pages (from-to)41-49
Number of pages9
JournalInformation Sciences
Volume441
DOIs
Publication statusPublished - 01-05-2018

Fingerprint

Computer aided diagnosis
Computer-aided Diagnosis
Convolution
Digital Image
Neural Networks
Neural networks
Nerve
Screening
Optics
High Accuracy
Robustness
Partial
Testing
Experimental Results
Demonstrate
Vision
Learning
Class
Deep learning

All Science Journal Classification (ASJC) codes

  • Software
  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computer Science Applications
  • Information Systems and Management
  • Artificial Intelligence

Cite this

@article{3cc7bfcf7f2540a2a5dd2d6438d84c67,
title = "Deep convolution neural network for accurate diagnosis of glaucoma using digital fundus images",
abstract = "Glaucoma progressively affects the optic nerve and may cause partial or complete vision loss. Raised intravascular pressure is the only factor which can be modified to prevent blindness from this condition. Accurate early detection and continuous screening may prevent the vision loss. Computer aided diagnosis (CAD) is a non-invasive technique which can detect the glaucoma in its early stage using digital fundus images. Developing such a system require diverse huge database in order to reach optimum performance. This paper proposes a novel CAD tool for the accurate detection of glaucoma using deep learning technique. An eighteen layer convolutional neural networks (CNN) is effectively trained in order to extract robust features from the digital fundus images. Finally these features are classified into normal and glaucoma classes during testing. We have achieved the highest accuracy of 98.13{\%} using 1426 (589: normal and 837: glaucoma) fundus images. Our experimental results demonstrates the robustness of the system, which can be used as a supplementary tool for the clinicians to validate their decisions.",
author = "U. Raghavendra and Hamido Fujita and Bhandary, {Sulatha V.} and Anjan Gudigar and Tan, {Jen Hong} and Acharya, {U. Rajendra}",
year = "2018",
month = "5",
day = "1",
doi = "10.1016/j.ins.2018.01.051",
language = "English",
volume = "441",
pages = "41--49",
journal = "Information Sciences",
issn = "0020-0255",
publisher = "Elsevier Inc.",

}

Deep convolution neural network for accurate diagnosis of glaucoma using digital fundus images. / Raghavendra, U.; Fujita, Hamido; Bhandary, Sulatha V.; Gudigar, Anjan; Tan, Jen Hong; Acharya, U. Rajendra.

In: Information Sciences, Vol. 441, 01.05.2018, p. 41-49.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Deep convolution neural network for accurate diagnosis of glaucoma using digital fundus images

AU - Raghavendra, U.

AU - Fujita, Hamido

AU - Bhandary, Sulatha V.

AU - Gudigar, Anjan

AU - Tan, Jen Hong

AU - Acharya, U. Rajendra

PY - 2018/5/1

Y1 - 2018/5/1

N2 - Glaucoma progressively affects the optic nerve and may cause partial or complete vision loss. Raised intravascular pressure is the only factor which can be modified to prevent blindness from this condition. Accurate early detection and continuous screening may prevent the vision loss. Computer aided diagnosis (CAD) is a non-invasive technique which can detect the glaucoma in its early stage using digital fundus images. Developing such a system require diverse huge database in order to reach optimum performance. This paper proposes a novel CAD tool for the accurate detection of glaucoma using deep learning technique. An eighteen layer convolutional neural networks (CNN) is effectively trained in order to extract robust features from the digital fundus images. Finally these features are classified into normal and glaucoma classes during testing. We have achieved the highest accuracy of 98.13% using 1426 (589: normal and 837: glaucoma) fundus images. Our experimental results demonstrates the robustness of the system, which can be used as a supplementary tool for the clinicians to validate their decisions.

AB - Glaucoma progressively affects the optic nerve and may cause partial or complete vision loss. Raised intravascular pressure is the only factor which can be modified to prevent blindness from this condition. Accurate early detection and continuous screening may prevent the vision loss. Computer aided diagnosis (CAD) is a non-invasive technique which can detect the glaucoma in its early stage using digital fundus images. Developing such a system require diverse huge database in order to reach optimum performance. This paper proposes a novel CAD tool for the accurate detection of glaucoma using deep learning technique. An eighteen layer convolutional neural networks (CNN) is effectively trained in order to extract robust features from the digital fundus images. Finally these features are classified into normal and glaucoma classes during testing. We have achieved the highest accuracy of 98.13% using 1426 (589: normal and 837: glaucoma) fundus images. Our experimental results demonstrates the robustness of the system, which can be used as a supplementary tool for the clinicians to validate their decisions.

UR - http://www.scopus.com/inward/record.url?scp=85042049995&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042049995&partnerID=8YFLogxK

U2 - 10.1016/j.ins.2018.01.051

DO - 10.1016/j.ins.2018.01.051

M3 - Article

AN - SCOPUS:85042049995

VL - 441

SP - 41

EP - 49

JO - Information Sciences

JF - Information Sciences

SN - 0020-0255

ER -