Automated detection of age-related macular degeneration using empirical mode decomposition

Muthu Rama Krishnan Mookiah, U. Rajendra Acharya, Hamido Fujita, Joel E.W. Koh, Jen Hong Tan, Chua Kuang Chua, Sulatha V. Bhandary, Kevin Noronha, Augustinus Laude, Louis Tong

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Age-related Macular Degeneration (AMD) is the posterior segment eye disease affecting elderly people and may lead to loss of vision. AMD is diagnosed using clinical features like drusen, Geographic Atrophy (GA) and Choroidal NeoVascularization (CNV) present in the fundus image. It is mainly classified into dry and wet type. Dry AMD is most common among elderly people. At present there is no treatment available for dry AMD. Early diagnosis and treatment to the affected eye may reduce the progression of disease. Manual screening of fundus images is time consuming and subjective. Hence in this study we are proposing an Empirical Mode Decomposition (EMD)-based nonlinear feature extraction to characterize and classify normal and AMD fundus images. EMD is performed on 1D Radon Transform (RT) projections to generate different Intrinsic Mode Functions (IMF). Various nonlinear features are extracted from the IMFs. The dimensionality of the extracted features are reduced using Locality Sensitive Discriminant Analysis (LSDA). Then the reduced LSDA features are ranked using minimum Redundancy Maximum Relevance (mRMR), Kullback-Leibler Divergence (KLD) and Chernoff Bound and Bhattacharyya Distance (CBBD) techniques. Ranked LSDA components are sequentially fed to Support Vector Machine (SVM) classifier to discriminate normal and AMD classes. The performance of the current study is experimented using private and two public datasets namely Automated Retinal Image Analysis (ARIA) and STructured Analysis of the Retina (STARE). The 10-fold cross validation approach is used to evaluate the performance of the classifiers and obtained highest average classification accuracy of 100%, sensitivity of 100% and specificity of 100% for STARE dataset using only two ranked LSDA components. Our results reveal that the proposed system can be used as a decision support tool for clinicians for mass AMD screening.

Original languageEnglish
Pages (from-to)654-668
Number of pages15
JournalKnowledge-Based Systems
Volume89
DOIs
Publication statusPublished - 01-01-2015

Fingerprint

Discriminant analysis
Decomposition
Screening
Classifiers
Radon
Image analysis
Support vector machines
Redundancy
Feature extraction
Mathematical transformations
Locality

All Science Journal Classification (ASJC) codes

  • Software
  • Management Information Systems
  • Information Systems and Management
  • Artificial Intelligence

Cite this

Mookiah, M. R. K., Acharya, U. R., Fujita, H., Koh, J. E. W., Tan, J. H., Chua, C. K., ... Tong, L. (2015). Automated detection of age-related macular degeneration using empirical mode decomposition. Knowledge-Based Systems, 89, 654-668. https://doi.org/10.1016/j.knosys.2015.09.012
Mookiah, Muthu Rama Krishnan ; Acharya, U. Rajendra ; Fujita, Hamido ; Koh, Joel E.W. ; Tan, Jen Hong ; Chua, Chua Kuang ; Bhandary, Sulatha V. ; Noronha, Kevin ; Laude, Augustinus ; Tong, Louis. / Automated detection of age-related macular degeneration using empirical mode decomposition. In: Knowledge-Based Systems. 2015 ; Vol. 89. pp. 654-668.
@article{9d4867b9f1a946338733cb46f14156ae,
title = "Automated detection of age-related macular degeneration using empirical mode decomposition",
abstract = "Age-related Macular Degeneration (AMD) is the posterior segment eye disease affecting elderly people and may lead to loss of vision. AMD is diagnosed using clinical features like drusen, Geographic Atrophy (GA) and Choroidal NeoVascularization (CNV) present in the fundus image. It is mainly classified into dry and wet type. Dry AMD is most common among elderly people. At present there is no treatment available for dry AMD. Early diagnosis and treatment to the affected eye may reduce the progression of disease. Manual screening of fundus images is time consuming and subjective. Hence in this study we are proposing an Empirical Mode Decomposition (EMD)-based nonlinear feature extraction to characterize and classify normal and AMD fundus images. EMD is performed on 1D Radon Transform (RT) projections to generate different Intrinsic Mode Functions (IMF). Various nonlinear features are extracted from the IMFs. The dimensionality of the extracted features are reduced using Locality Sensitive Discriminant Analysis (LSDA). Then the reduced LSDA features are ranked using minimum Redundancy Maximum Relevance (mRMR), Kullback-Leibler Divergence (KLD) and Chernoff Bound and Bhattacharyya Distance (CBBD) techniques. Ranked LSDA components are sequentially fed to Support Vector Machine (SVM) classifier to discriminate normal and AMD classes. The performance of the current study is experimented using private and two public datasets namely Automated Retinal Image Analysis (ARIA) and STructured Analysis of the Retina (STARE). The 10-fold cross validation approach is used to evaluate the performance of the classifiers and obtained highest average classification accuracy of 100{\%}, sensitivity of 100{\%} and specificity of 100{\%} for STARE dataset using only two ranked LSDA components. Our results reveal that the proposed system can be used as a decision support tool for clinicians for mass AMD screening.",
author = "Mookiah, {Muthu Rama Krishnan} and Acharya, {U. Rajendra} and Hamido Fujita and Koh, {Joel E.W.} and Tan, {Jen Hong} and Chua, {Chua Kuang} and Bhandary, {Sulatha V.} and Kevin Noronha and Augustinus Laude and Louis Tong",
year = "2015",
month = "1",
day = "1",
doi = "10.1016/j.knosys.2015.09.012",
language = "English",
volume = "89",
pages = "654--668",
journal = "Knowledge-Based Systems",
issn = "0950-7051",
publisher = "Elsevier",

}

Mookiah, MRK, Acharya, UR, Fujita, H, Koh, JEW, Tan, JH, Chua, CK, Bhandary, SV, Noronha, K, Laude, A & Tong, L 2015, 'Automated detection of age-related macular degeneration using empirical mode decomposition', Knowledge-Based Systems, vol. 89, pp. 654-668. https://doi.org/10.1016/j.knosys.2015.09.012

Automated detection of age-related macular degeneration using empirical mode decomposition. / Mookiah, Muthu Rama Krishnan; Acharya, U. Rajendra; Fujita, Hamido; Koh, Joel E.W.; Tan, Jen Hong; Chua, Chua Kuang; Bhandary, Sulatha V.; Noronha, Kevin; Laude, Augustinus; Tong, Louis.

In: Knowledge-Based Systems, Vol. 89, 01.01.2015, p. 654-668.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Automated detection of age-related macular degeneration using empirical mode decomposition

AU - Mookiah, Muthu Rama Krishnan

AU - Acharya, U. Rajendra

AU - Fujita, Hamido

AU - Koh, Joel E.W.

AU - Tan, Jen Hong

AU - Chua, Chua Kuang

AU - Bhandary, Sulatha V.

AU - Noronha, Kevin

AU - Laude, Augustinus

AU - Tong, Louis

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Age-related Macular Degeneration (AMD) is the posterior segment eye disease affecting elderly people and may lead to loss of vision. AMD is diagnosed using clinical features like drusen, Geographic Atrophy (GA) and Choroidal NeoVascularization (CNV) present in the fundus image. It is mainly classified into dry and wet type. Dry AMD is most common among elderly people. At present there is no treatment available for dry AMD. Early diagnosis and treatment to the affected eye may reduce the progression of disease. Manual screening of fundus images is time consuming and subjective. Hence in this study we are proposing an Empirical Mode Decomposition (EMD)-based nonlinear feature extraction to characterize and classify normal and AMD fundus images. EMD is performed on 1D Radon Transform (RT) projections to generate different Intrinsic Mode Functions (IMF). Various nonlinear features are extracted from the IMFs. The dimensionality of the extracted features are reduced using Locality Sensitive Discriminant Analysis (LSDA). Then the reduced LSDA features are ranked using minimum Redundancy Maximum Relevance (mRMR), Kullback-Leibler Divergence (KLD) and Chernoff Bound and Bhattacharyya Distance (CBBD) techniques. Ranked LSDA components are sequentially fed to Support Vector Machine (SVM) classifier to discriminate normal and AMD classes. The performance of the current study is experimented using private and two public datasets namely Automated Retinal Image Analysis (ARIA) and STructured Analysis of the Retina (STARE). The 10-fold cross validation approach is used to evaluate the performance of the classifiers and obtained highest average classification accuracy of 100%, sensitivity of 100% and specificity of 100% for STARE dataset using only two ranked LSDA components. Our results reveal that the proposed system can be used as a decision support tool for clinicians for mass AMD screening.

AB - Age-related Macular Degeneration (AMD) is the posterior segment eye disease affecting elderly people and may lead to loss of vision. AMD is diagnosed using clinical features like drusen, Geographic Atrophy (GA) and Choroidal NeoVascularization (CNV) present in the fundus image. It is mainly classified into dry and wet type. Dry AMD is most common among elderly people. At present there is no treatment available for dry AMD. Early diagnosis and treatment to the affected eye may reduce the progression of disease. Manual screening of fundus images is time consuming and subjective. Hence in this study we are proposing an Empirical Mode Decomposition (EMD)-based nonlinear feature extraction to characterize and classify normal and AMD fundus images. EMD is performed on 1D Radon Transform (RT) projections to generate different Intrinsic Mode Functions (IMF). Various nonlinear features are extracted from the IMFs. The dimensionality of the extracted features are reduced using Locality Sensitive Discriminant Analysis (LSDA). Then the reduced LSDA features are ranked using minimum Redundancy Maximum Relevance (mRMR), Kullback-Leibler Divergence (KLD) and Chernoff Bound and Bhattacharyya Distance (CBBD) techniques. Ranked LSDA components are sequentially fed to Support Vector Machine (SVM) classifier to discriminate normal and AMD classes. The performance of the current study is experimented using private and two public datasets namely Automated Retinal Image Analysis (ARIA) and STructured Analysis of the Retina (STARE). The 10-fold cross validation approach is used to evaluate the performance of the classifiers and obtained highest average classification accuracy of 100%, sensitivity of 100% and specificity of 100% for STARE dataset using only two ranked LSDA components. Our results reveal that the proposed system can be used as a decision support tool for clinicians for mass AMD screening.

UR - http://www.scopus.com/inward/record.url?scp=84944316424&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944316424&partnerID=8YFLogxK

U2 - 10.1016/j.knosys.2015.09.012

DO - 10.1016/j.knosys.2015.09.012

M3 - Article

AN - SCOPUS:84944316424

VL - 89

SP - 654

EP - 668

JO - Knowledge-Based Systems

JF - Knowledge-Based Systems

SN - 0950-7051

ER -