A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity

V. Bolon-Canedo, E. Ataer-Cansizoglu, D. Erdogmus, Jayashree Kalpathy-Cramer, Michael Chiang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Retinopathy of Prematurity (ROP) is an ophthalmic disease that is a leading cause of childhood blindness throughout the world. Accurate diagnosis of ROP is vital to identify infants who require treatment, which can prevent blindness. Arterial tortuosity and venous dilation in the retina are important signs of ROP, so it is necessary to extract these features from points on the vessels or vessel segments. Then, an image is represented with statistics such as minimum, maximum or mean of these values. However, these statistics provide biased estimates as an image contains both healthy and abnormal vessels. In this work, we present a novel feature extraction technique that represents each image with the parameters of a two-component Gaussian Mixture Model (GMM). Using these features, we performed classification experiments on a manually segmented retinal image dataset consisting of 77 images. The results show that GMM-based features outperform other features that are based on classical statistics, with accuracy over 90%. Moreover, if the features are extracted from the whole image without distinguishing veins and arteries, proposed features provide better performance compared to using traditional statistics.

Original languageEnglish (US)
Title of host publicationProceedings - International Symposium on Biomedical Imaging
PublisherIEEE Computer Society
Pages1498-1501
Number of pages4
Volume2015-July
ISBN (Print)9781479923748
DOIs
StatePublished - Jul 21 2015
Event12th IEEE International Symposium on Biomedical Imaging, ISBI 2015 - Brooklyn, United States
Duration: Apr 16 2015Apr 19 2015

Other

Other12th IEEE International Symposium on Biomedical Imaging, ISBI 2015
CountryUnited States
CityBrooklyn
Period4/16/154/19/15

Fingerprint

Retinopathy of Prematurity
Feature extraction
Statistics
Blindness
Eye Diseases
Retina
Dilatation
Veins
Arteries
Experiments
Therapeutics

Keywords

  • classification
  • feature extraction
  • Retinopathy of prematurity

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Bolon-Canedo, V., Ataer-Cansizoglu, E., Erdogmus, D., Kalpathy-Cramer, J., & Chiang, M. (2015). A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity. In Proceedings - International Symposium on Biomedical Imaging (Vol. 2015-July, pp. 1498-1501). [7164161] IEEE Computer Society. https://doi.org/10.1109/ISBI.2015.7164161

A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity. / Bolon-Canedo, V.; Ataer-Cansizoglu, E.; Erdogmus, D.; Kalpathy-Cramer, Jayashree; Chiang, Michael.

Proceedings - International Symposium on Biomedical Imaging. Vol. 2015-July IEEE Computer Society, 2015. p. 1498-1501 7164161.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Bolon-Canedo, V, Ataer-Cansizoglu, E, Erdogmus, D, Kalpathy-Cramer, J & Chiang, M 2015, A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity. in Proceedings - International Symposium on Biomedical Imaging. vol. 2015-July, 7164161, IEEE Computer Society, pp. 1498-1501, 12th IEEE International Symposium on Biomedical Imaging, ISBI 2015, Brooklyn, United States, 4/16/15. https://doi.org/10.1109/ISBI.2015.7164161
Bolon-Canedo V, Ataer-Cansizoglu E, Erdogmus D, Kalpathy-Cramer J, Chiang M. A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity. In Proceedings - International Symposium on Biomedical Imaging. Vol. 2015-July. IEEE Computer Society. 2015. p. 1498-1501. 7164161 https://doi.org/10.1109/ISBI.2015.7164161
Bolon-Canedo, V. ; Ataer-Cansizoglu, E. ; Erdogmus, D. ; Kalpathy-Cramer, Jayashree ; Chiang, Michael. / A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity. Proceedings - International Symposium on Biomedical Imaging. Vol. 2015-July IEEE Computer Society, 2015. pp. 1498-1501
@inproceedings{90bec6050ce4499cb46d1ac79a6ede34,
title = "A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity",
abstract = "Retinopathy of Prematurity (ROP) is an ophthalmic disease that is a leading cause of childhood blindness throughout the world. Accurate diagnosis of ROP is vital to identify infants who require treatment, which can prevent blindness. Arterial tortuosity and venous dilation in the retina are important signs of ROP, so it is necessary to extract these features from points on the vessels or vessel segments. Then, an image is represented with statistics such as minimum, maximum or mean of these values. However, these statistics provide biased estimates as an image contains both healthy and abnormal vessels. In this work, we present a novel feature extraction technique that represents each image with the parameters of a two-component Gaussian Mixture Model (GMM). Using these features, we performed classification experiments on a manually segmented retinal image dataset consisting of 77 images. The results show that GMM-based features outperform other features that are based on classical statistics, with accuracy over 90{\%}. Moreover, if the features are extracted from the whole image without distinguishing veins and arteries, proposed features provide better performance compared to using traditional statistics.",
keywords = "classification, feature extraction, Retinopathy of prematurity",
author = "V. Bolon-Canedo and E. Ataer-Cansizoglu and D. Erdogmus and Jayashree Kalpathy-Cramer and Michael Chiang",
year = "2015",
month = "7",
day = "21",
doi = "10.1109/ISBI.2015.7164161",
language = "English (US)",
isbn = "9781479923748",
volume = "2015-July",
pages = "1498--1501",
booktitle = "Proceedings - International Symposium on Biomedical Imaging",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - A GMM-based feature extraction technique for the automated diagnosis of Retinopathy of Prematurity

AU - Bolon-Canedo, V.

AU - Ataer-Cansizoglu, E.

AU - Erdogmus, D.

AU - Kalpathy-Cramer, Jayashree

AU - Chiang, Michael

PY - 2015/7/21

Y1 - 2015/7/21

N2 - Retinopathy of Prematurity (ROP) is an ophthalmic disease that is a leading cause of childhood blindness throughout the world. Accurate diagnosis of ROP is vital to identify infants who require treatment, which can prevent blindness. Arterial tortuosity and venous dilation in the retina are important signs of ROP, so it is necessary to extract these features from points on the vessels or vessel segments. Then, an image is represented with statistics such as minimum, maximum or mean of these values. However, these statistics provide biased estimates as an image contains both healthy and abnormal vessels. In this work, we present a novel feature extraction technique that represents each image with the parameters of a two-component Gaussian Mixture Model (GMM). Using these features, we performed classification experiments on a manually segmented retinal image dataset consisting of 77 images. The results show that GMM-based features outperform other features that are based on classical statistics, with accuracy over 90%. Moreover, if the features are extracted from the whole image without distinguishing veins and arteries, proposed features provide better performance compared to using traditional statistics.

AB - Retinopathy of Prematurity (ROP) is an ophthalmic disease that is a leading cause of childhood blindness throughout the world. Accurate diagnosis of ROP is vital to identify infants who require treatment, which can prevent blindness. Arterial tortuosity and venous dilation in the retina are important signs of ROP, so it is necessary to extract these features from points on the vessels or vessel segments. Then, an image is represented with statistics such as minimum, maximum or mean of these values. However, these statistics provide biased estimates as an image contains both healthy and abnormal vessels. In this work, we present a novel feature extraction technique that represents each image with the parameters of a two-component Gaussian Mixture Model (GMM). Using these features, we performed classification experiments on a manually segmented retinal image dataset consisting of 77 images. The results show that GMM-based features outperform other features that are based on classical statistics, with accuracy over 90%. Moreover, if the features are extracted from the whole image without distinguishing veins and arteries, proposed features provide better performance compared to using traditional statistics.

KW - classification

KW - feature extraction

KW - Retinopathy of prematurity

UR - http://www.scopus.com/inward/record.url?scp=84944321422&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944321422&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2015.7164161

DO - 10.1109/ISBI.2015.7164161

M3 - Conference contribution

AN - SCOPUS:84944321422

SN - 9781479923748

VL - 2015-July

SP - 1498

EP - 1501

BT - Proceedings - International Symposium on Biomedical Imaging

PB - IEEE Computer Society

ER -