Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features

Devvi Sarwinda, Ari Wibisono, Hanifa Arrumaisha, Zaki Raihan, Rosa N. Rizky FT, Rico Putra Pradana, Mohammad Aulia Hafidh, Petrus Mursanto

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

Abstract

Diabetic retinopathy is a disease caused by the complications of diabetes mellitus and can cause blindness. In this study, we classified the stages of diabetic retinopathy using a large-scale dataset that consists of 35,126 fundus images. The classification of diabetic retinopathy includes five stages, from normal to proliferative diabetic retinopathy. In the proposed approach, a morphological feature extraction method and advanced local binary patterns were employed to extract blood vessel and texture features, respectively. The support vector machine, K-nearest neighbor, random forest, and logistic regression techniques were compared as classifiers. The classification was conducted on fundus images from a Kaggle dataset. The experimental results show that the texture feature extraction method based on advanced local binary patterns leads to higher accuracy, precision, and recall score than the blood vessel features extracted using morphological operators.

Original languageEnglish
Title of host publicationStudies in Computational Intelligence
PublisherSpringer Verlag
Pages141-155
Number of pages15
DOIs
Publication statusPublished - 1 Jan 2020

Publication series

NameStudies in Computational Intelligence
Volume850
ISSN (Print)1860-949X
ISSN (Electronic)1860-9503

Fingerprint

Blood vessels
Feature extraction
Textures
Medical problems
Support vector machines
Classifiers

Keywords

  • Blood vessel
  • Diabetic retinopathy
  • Fundus images
  • Large scale dataset
  • Texture feature

Cite this

Sarwinda, D., Wibisono, A., Arrumaisha, H., Raihan, Z., Rizky FT, R. N., Pradana, R. P., ... Mursanto, P. (2020). Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features. In Studies in Computational Intelligence (pp. 141-155). (Studies in Computational Intelligence; Vol. 850). Springer Verlag. https://doi.org/10.1007/978-3-030-26428-4_10
Sarwinda, Devvi ; Wibisono, Ari ; Arrumaisha, Hanifa ; Raihan, Zaki ; Rizky FT, Rosa N. ; Pradana, Rico Putra ; Hafidh, Mohammad Aulia ; Mursanto, Petrus. / Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features. Studies in Computational Intelligence. Springer Verlag, 2020. pp. 141-155 (Studies in Computational Intelligence).
@inbook{c756a37efe1c4382b9d8178e85e40d68,
title = "Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features",
abstract = "Diabetic retinopathy is a disease caused by the complications of diabetes mellitus and can cause blindness. In this study, we classified the stages of diabetic retinopathy using a large-scale dataset that consists of 35,126 fundus images. The classification of diabetic retinopathy includes five stages, from normal to proliferative diabetic retinopathy. In the proposed approach, a morphological feature extraction method and advanced local binary patterns were employed to extract blood vessel and texture features, respectively. The support vector machine, K-nearest neighbor, random forest, and logistic regression techniques were compared as classifiers. The classification was conducted on fundus images from a Kaggle dataset. The experimental results show that the texture feature extraction method based on advanced local binary patterns leads to higher accuracy, precision, and recall score than the blood vessel features extracted using morphological operators.",
keywords = "Blood vessel, Diabetic retinopathy, Fundus images, Large scale dataset, Texture feature",
author = "Devvi Sarwinda and Ari Wibisono and Hanifa Arrumaisha and Zaki Raihan and {Rizky FT}, {Rosa N.} and Pradana, {Rico Putra} and Hafidh, {Mohammad Aulia} and Petrus Mursanto",
year = "2020",
month = "1",
day = "1",
doi = "10.1007/978-3-030-26428-4_10",
language = "English",
series = "Studies in Computational Intelligence",
publisher = "Springer Verlag",
pages = "141--155",
booktitle = "Studies in Computational Intelligence",
address = "Germany",

}

Sarwinda, D, Wibisono, A, Arrumaisha, H, Raihan, Z, Rizky FT, RN, Pradana, RP, Hafidh, MA & Mursanto, P 2020, Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features. in Studies in Computational Intelligence. Studies in Computational Intelligence, vol. 850, Springer Verlag, pp. 141-155. https://doi.org/10.1007/978-3-030-26428-4_10

Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features. / Sarwinda, Devvi; Wibisono, Ari; Arrumaisha, Hanifa; Raihan, Zaki; Rizky FT, Rosa N.; Pradana, Rico Putra; Hafidh, Mohammad Aulia; Mursanto, Petrus.

Studies in Computational Intelligence. Springer Verlag, 2020. p. 141-155 (Studies in Computational Intelligence; Vol. 850).

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

TY - CHAP

T1 - Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features

AU - Sarwinda, Devvi

AU - Wibisono, Ari

AU - Arrumaisha, Hanifa

AU - Raihan, Zaki

AU - Rizky FT, Rosa N.

AU - Pradana, Rico Putra

AU - Hafidh, Mohammad Aulia

AU - Mursanto, Petrus

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Diabetic retinopathy is a disease caused by the complications of diabetes mellitus and can cause blindness. In this study, we classified the stages of diabetic retinopathy using a large-scale dataset that consists of 35,126 fundus images. The classification of diabetic retinopathy includes five stages, from normal to proliferative diabetic retinopathy. In the proposed approach, a morphological feature extraction method and advanced local binary patterns were employed to extract blood vessel and texture features, respectively. The support vector machine, K-nearest neighbor, random forest, and logistic regression techniques were compared as classifiers. The classification was conducted on fundus images from a Kaggle dataset. The experimental results show that the texture feature extraction method based on advanced local binary patterns leads to higher accuracy, precision, and recall score than the blood vessel features extracted using morphological operators.

AB - Diabetic retinopathy is a disease caused by the complications of diabetes mellitus and can cause blindness. In this study, we classified the stages of diabetic retinopathy using a large-scale dataset that consists of 35,126 fundus images. The classification of diabetic retinopathy includes five stages, from normal to proliferative diabetic retinopathy. In the proposed approach, a morphological feature extraction method and advanced local binary patterns were employed to extract blood vessel and texture features, respectively. The support vector machine, K-nearest neighbor, random forest, and logistic regression techniques were compared as classifiers. The classification was conducted on fundus images from a Kaggle dataset. The experimental results show that the texture feature extraction method based on advanced local binary patterns leads to higher accuracy, precision, and recall score than the blood vessel features extracted using morphological operators.

KW - Blood vessel

KW - Diabetic retinopathy

KW - Fundus images

KW - Large scale dataset

KW - Texture feature

UR - http://www.scopus.com/inward/record.url?scp=85071527244&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-26428-4_10

DO - 10.1007/978-3-030-26428-4_10

M3 - Chapter

T3 - Studies in Computational Intelligence

SP - 141

EP - 155

BT - Studies in Computational Intelligence

PB - Springer Verlag

ER -

Sarwinda D, Wibisono A, Arrumaisha H, Raihan Z, Rizky FT RN, Pradana RP et al. Analysis of Large-Scale Diabetic Retinopathy Datasets Using Texture and Blood Vessel Features. In Studies in Computational Intelligence. Springer Verlag. 2020. p. 141-155. (Studies in Computational Intelligence). https://doi.org/10.1007/978-3-030-26428-4_10